47938 DISTORTIONS TOAGRICULTURAL INCENTIVES ASIA IN Editors KymAnderson ·Will Martin Distortions to Agricultural Incentives in Asia Distortions to Agricultural Incentives in Asia Kym Anderson and Will Martin, Editors Washington, D.C. © 2009 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved. 1 2 3 4 12 11 10 09 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without per- mission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Pub- lisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Cover design: Tomoko Hirata/World Bank. Cover photo: © Tran Thi Hoa/World Bank Photo Library. ISBN: 978-0-8213-7662-1 eISBN: 978-0-8213-7663-8 DOI: 10.1596/978-0-8213-7662-1 Library of Congress Cataloging-in-Publication Data Distortions to agricultural incentives in Asia / edited by Kym Anderson and Will Martin. p. cm. Includes bibliographical references and index. ISBN 978-0-8213-7662-1 -- ISBN 978-0-8213-7663-8 (electronic) 1. Agriculture--Economic aspects--Asia. 2. Agriculture and state--Asia. 3. Agriculture--Taxation--Asia. 4. Agricultural subsidies--Asia. I. Anderson, Kym. II. Martin, Will, 1953- HD2056.Z8D57 2008 338.1'85--dc22 2008029534 Dedication To the authors of the country case studies and their assistants, especially for generating the time series of distortion estimates that underpin the chapters, and, in particular, to Yujiro Hayami for his insights and advice during this project and his related and influential work on Asia over several decades. CONTENTS Foreword xvii Acknowledgments xxi Contributors xxiii Abbreviations xxvii Map: The Focus Economies of Asia xxviii PART I INTRODUCTION 1 1 Introduction and Summary 3 Kym Anderson and Will Martin PART II NORTHEAST ASIA 83 2 Republic of Korea and Taiwan, China 85 Masayoshi Honma and Yujiro Hayami 3 China 117 Jikun Huang, Scott Rozelle, Will Martin, and Yu Liu PART III SOUTHEAST ASIA 163 4 Indonesia 165 George Fane and Peter Warr 5 Malaysia 197 Prema-Chandra Athukorala and Wai-Heng Loke 6 The Philippines 223 Cristina David, Ponciano Intal, and Arsenio M. Balisacan vii viii Contents 7 Thailand 255 Peter Warr and Archanun Kohpaiboon 8 Vietnam 281 Prema-Chandra Athukorala, Pham Lan Huong, and Vo Tri Thanh PART IV SOUTH ASIA 303 9 Bangladesh 305 Nazneen Ahmed, Zaid Bakht, Paul A. Dorosh, and Quazi Shahabuddin 10 India 339 Garry Pursell, Ashok Gulati, and Kanupriya Gupta 11 Pakistan 379 Paul A. Dorosh and Abdul Salam 12 Sri Lanka 409 Jayatilleke Bandara and Sisira Jayasuriya Appendix A: Methodology for Measuring Distortions to Agricultural Incentives 441 Kym Anderson, Marianne Kurzweil, Will Martin, Damiano Sandri, and Ernesto Valenzuela Appendix B: Annual Estimates of Asian Distortions to Agricultural Incentives 473 Ernesto Valenzuela, Marianne Kurzweil, Johanna Croser, Signe Nelgen, and Kym Anderson Index 563 Figures 1.1 Index of Real Per Capita GDP, Asia Relative to the United States, 1950­2006 11 1.2 NRAs in Agriculture, Asian Focus Economies, 1980­84 and 2000­04 28 1.3 NRAs, by Product, Asian Focus Economies, 1980­84 and 2000­04 29 1.4 NRAs for Rice, Milk, and Sugar, Asian Focus Economies, 1980­84 and 2000­04 30 1.5 NRAs for Exportable, Import-Competing, and All Agricultural Products, Asian Focus Economies, 1955­2004 33 1.6 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Asian Focus Economies, 1955­2004 51 Contents ix 1.7 RRAs in Agriculture, Asian Focus Economies, 1980­84 and 2000­04 52 1.8 NRAs and RRAs, China and India, 1965­2004 53 1.9 NRAs and RRAs, Africa, Asia, and Latin America, 1965­2004 54 1.10 Income Distribution, Asian Subregions and the World, 2000 55 1.11 Regressions of Real GDP Per Capita and Agricultural NRAs and RRAs, Asian Focus Economies, 1955­2005 58 1.12 Regressions of Real Comparative Advantage and Agricultural NRAs and RRAs, Asian Focus Economies, 1960­2004 59 1.13 NRAs and International Prices, Rice, Asian Focus Economies, 1970­2005 69 1.14 NRAs for Rice, Malaysia, 1960­2004 71 1.15 RRAs and TBIs in Agriculture, Asian Focus Economies, 1980­84 and 2000­04 72 1.16 RRAs and Real Per Capita GDP, India, Japan, and Northeast Asian Focus Economies, 1955­2005 75 1.17 NRAs for China, Japan, and the Republic of Korea and GATT/WTO Accession, 1955­2005 76 2.1 RRAs for Agricultural and Nonagricultural Tradables, Japan, Republic of Korea, and Taiwan, China, 1955­2004 103 2.2 NRAs for Agricultural Products, Republic of Korea and Taiwan, China, 1955­2004 104 2.3 NRAs for Exportable, Import-Competing, and All Agricultural Products, Taiwan, China, 1955­2002 105 2.4 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Republic of Korea and Taiwan, China, 1955­2004 106 2.5 NRAs for Rice, Republic of Korea and Taiwan, China, 1955­2004 107 2.6 RRAs in Agriculture and Relative GDP per Agricultural Worker, Republic of Korea and Taiwan, China, 1955­2004 112 3.1 Agricultural Trade Balance, by Factor Intensity, China, 1985­2002 128 3.2 The Domestic Market for Foreign Currency, China 141 3.3 NRAs for Exportable, Import-Competing, and All Agricultural Products, China, 1981­2004 149 3.4 Average NRAs for Producers of Major Commodities, China, 2000­05 151 3.5 NRA Estimates in Other Studies, by Product, China, 1995­2001 154 3.6 NRAs for Agricultural and Nonagricultural Tradables and the RRA, China, 1981­2004 155 4.1 Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Rice and Urea Fertilizer, Indonesia, 1975­2005 174 4.2 Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Sugar, Indonesia, 1971­2005 177 x Contents 4.3 Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Soybeans, Indonesia, 1970­2005 178 4.4 Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Maize, Indonesia, 1969­2005 179 4.5 Border and Domestic Prices of Export Crops Relative to the GDP Deflator, Indonesia, 1967­2005 181 4.6 NRAs for Exportable, Import-Competing, and All Agricultural Products, Indonesia, 1970­2004 192 4.7 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Indonesia, 1970­2004 194 5.1 Commodity Composition of Agricultural GDP, Malaysia, 1965­2005 200 5.2 NRAs for Exportable, Import-Competing, and All Agricultural Products, Malaysia, 1960­2004 215 5.3 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Malaysia, 1960­2004 217 6.1 Value Shares of the Primary Production of Covered and Noncovered Commodities, the Philippines, 1966­2004 239 6.2 NRAs for Exportable, Import-Competing, and All Agricultural Products, the Philippines, 1962­2004 245 6.3 NRAs for Agricultural and Nonagricultural Tradables and the RRA, the Philippines, 1962­2004 248 7.1 Price Comparisons and NRPs at Wholesale, Rice, Thailand, 1968­2005 262 7.2 Price Comparisons and NRPs at Wholesale, Maize, Thailand, 1968­2005 263 7.3 Price Comparisons and NRPs at Wholesale, Cassava, Thailand, 1969­2004 264 7.4 Price Comparisons and NRPs at Wholesale, Soybeans, Thailand, 1984­2005 265 7.5 Ratios, Consumer Prices to Border Prices and Miller Prices to Grower Prices, Sugar, Thailand, 1968­2005 266 7.6 Price Comparisons and NRPs at Wholesale, Sugar, Thailand, 1968­2005 267 7.7 Price Comparisons and NRPs at Wholesale, Palm Oil, Thailand, 1995­2004 267 7.8 Price Comparisons and NRPs at Wholesale, Rubber, Thailand, 1968­2005 268 7.9 Price Comparisons and NRPs at Wholesale, Urea Fertilizer, Thailand, 1984­2005 269 7.10 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Thailand, 1970­2004 277 8.1 Agriculture, GDP, and Value Added, Vietnam, 1985­2005 284 Contents xi 8.2 Commodity Shares in Agricultural Production, Vietnam, 1991­2002 286 8.3 Volume, Value, and Price Indexes, Agricultural Exports, Vietnam, 1990­2004 288 8.4 Weighted Average Import Duties, Vietnam, 1990­2004 292 8.5 Index, Real Exchange Rate, Vietnam, 1988­2005 294 8.6 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Vietnam, 1986­2004 298 9.1 Commodity Shares in Agricultural Production, Bangladesh, 1971­2003 315 9.2 NRAs and Border Prices, Rice, Bangladesh, 1974­2004 317 9.3 Prices and Private Sector Imports, Rice, Bangladesh, 1997­2007 319 9.4 NRAs for Exportable, Import-Competing, and All Agricultural Products, Bangladesh, 1974­2004 326 9.5 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Bangladesh, 1974­2004 328 10.1 Import Tariffs, Agricultural and Nonagricultural Products, India, 2002­06 347 10.2 Index, Real Effective Exchange Rate, India, 1965­2004 350 10.3 NRAs for Exportable, Import-Competing, and All Agricultural Products, India, 1965­2004 360 10.4 Real Domestic Producer Prices and International Reference Prices, Rice, India, 1965­2004 363 10.5 NRAs for Agricultural and Nonagricultural Tradables and the RRA, India, 1965­2004 366 11.1 Commodity Shares in Agricultural Production, Pakistan, 1965­2005 381 11.2 Import Tariffs and the Real Exchange Rate, Pakistan, 1985­2004 385 11.3 NRAs for Major Covered Products, Pakistan, 1962­2005 396 11.4 NRAs for Exportable, Import-Competing, and All Agricultural Products, Pakistan, 1973­2005 398 11.5 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Pakistan, 1973­2005 401 12.1 Real GDP Growth, Political Episodes, and Policy Regimes, Sri Lanka, 1951­2005 411 12.2 Agriculture in GDP and Exports, Sri Lanka, 1950­2005 414 12.3 Commodity Shares in Agricultural Production, Sri Lanka, 1966­2004 414 12.4 Fertilizer Subsidy, Sri Lanka, 1962­2007 419 12.5 NRAs for Tea, Sri Lanka, 1955­2004 423 12.6 NRAs for Rubber, Sri Lanka, 1955­2004 425 12.7 NRAs for Coconuts, Sri Lanka, 1955­2004 426 12.8 NRAs for Rice, Sri Lanka, 1955­2004 426 xii Contents 12.9 NRAs for Exportable, Import-Competing, and All Agricultural Products, Sri Lanka, 1955­2004 428 12.10 NRAs for Agricultural and Nonagricultural Tradables and the RRA, Sri Lanka, 1955­2004 429 A.1 A Distorted Domestic Market for Foreign Currency 445 A.2 Distorted Domestic Markets for Farm Products 464 Tables 1.1 Key Economic and Trade Indicators, Asian Focus Economies, 2000­04 6 1.2 Poverty Levels in Asia, 1981­2004 8 1.3 Real Growth in GDP and Exports, Asian Focus Economies, 1980­2004 10 1.4 Exports of Goods and Services as a Share of GDP, Asian Focus Economies, 1965­2004 12 1.5 Share of Nonfood Manufactures in World Exports, Asian Focus Economies, 1990­2006 13 1.6 Sectoral Shares of GDP, Asian Focus Economies, 1965­2004 14 1.7 Agriculture's Share in Employment, Asian Focus Economies, 1965­2004 15 1.8 Sectoral Shares of Merchandise Exports, Asian Focus Economies, 1965­2004 16 1.9 Indexes of Comparative Advantage in Agriculture and Processed Food, Asian Focus Economies, 1965­2004 18 1.10 Export Orientation, Import Dependence, and Self-Sufficiency in Primary Agricultural Production, Asian Focus Economies, 1961­2004 19 1.11 NRAs for Agricultural Products, Asian Focus Economies, 1955­2004 27 1.12 NRA Dispersion across Covered Agricultural Products, Asian Focus Economies, 1955­2004 32 1.13 NRAs for Agricultural Exportable and Import-Competing Products and the TBI, Asian Focus Economies, 1955­2004 34 1.14 NRAs for Covered Agricultural Products, by Policy Instrument, Asian Focus Economies, 1955­2004 38 1.15 Gross Subsidy Equivalents of Agricultural Assistance, Total and Per Farmworker, Asian Focus Economies, 1955­2004 41 1.16 NRA and Gross Subsidy Equivalents of Farmer Assistance, by Product, Asian Focus Economies, 1955­2004 45 1.17 RRAs in Agriculture, Asian Focus Economies, 1955­2004 48 1.18 Relative Per Capita Income, Agricultural Comparative Advantage, and NRAs and RRAs for Agricultural Tradables, Asian Focus Economies, 2000­04 57 Contents xiii 1.19 Regressions of NRAs and Selected Determinants, Asian Focus Economies, 1960­2004 61 1.20 CTEs for Policies Assisting the Producers of Covered Agricultural Products, Asian Focus Economies, 1960­2004 62 1.21 Value of CTEs for Policies Assisting the Producers of Covered Agricultural Products, Asian Focus Economies, 1965­2004 65 2.1 Economic Growth and Structural Transformation, Republic of Korea and Taiwan, China, 1955­2004 89 2.2 Changes in Agricultural Structure, Republic of Korea and Taiwan, China, 1955­2004 92 2.3 NRAs for Covered Agricultural Products, Republic of Korea, 1955­2004 100 2.4 NRAs for Covered Agricultural Products, Taiwan, China, 1955­2002 101 2.5 NRAs in Agriculture Relative to Nonagricultural Industries, Republic of Korea and Taiwan, China, 1955­2004 102 2.6 CTEs for Covered Agricultural Products, Republic of Korea and Taiwan, China, 1955­2004 108 3.1 Real Average Annual Rates of Economic Growth, China, 1970­2004 126 3.2 Structure of the Agricultural Economy, China, 1970­2005 127 3.3 Rural Income Per Capita, China, 1980­2001 130 3.4 NRAs for Covered Agricultural Products, China, 1981­2005 145 3.5 NRAs in Agriculture Relative to Nonagricultural Industries, China, 1981­2005 152 3.6 CTEs for Covered Agricultural Products, China, 1981­2005 153 4.1 Real GDP Growth and Its Sectoral Components, Indonesia, 1968­2005 167 4.2 Subsector Shares of Agricultural Value Added, Indonesia, 1971­2000 168 4.3 Export Sales in Total Sales and Imports in Total Usage, Indonesia, 1971­2000 169 4.4 Estimated Effective Rates of Protection, Indonesia, 1987, 1995, and 2003 171 4.5 Estimates of Transmission Elasticities from Wholesale to Farm Prices, Indonesia 189 4.6 NRAs for Covered Agricultural Products, Including Fertilizer Use Subsidies, Indonesia, 1970­2004 191 4.7 NRAs in Agriculture Relative to Nonagricultural Industries, Indonesia, 1970­2004 193 5.1 Agriculture in the Economy, Malaysia, 1970­2005 200 5.2 Product Shares, Agricultural Exports, Malaysia, 1970­2004 205 5.3 NRAs for Covered Agricultural Products, Malaysia, 1960­2004 214 5.4 NRAs in Agriculture Relative to Nonagricultural Industries, Malaysia, 1960­2004 216 xiv Contents 6.1 Agriculture in the Economy, Growth Rates, the Philippines and Other Asian Countries, 1960­2004 224 6.2 Changes in Agricultural Structure and Trade Openness, the Philippines, 1960­2004 225 6.3 GVA Growth Rates, Major Agricultural Commodities, the Philippines, 1960­2004 227 6.4 Revealed Comparative Advantage, Major Agricultural Commodities, the Philippines, 1960­2004 229 6.5 NRAs for Covered Agricultural Products, the Philippines, 1962­2004 242 6.6 NRAs in Agriculture Relative to Nonagricultural Industries, the Philippines, 1962­2004 244 6.7 Coefficients of Variation in Real International Prices and Philippine Wholesale Prices, Major Agricultural Commodities, 1960­2004 246 7.1 Real GDP Growth and Its Sectoral Components, Thailand, 1968­2005 258 7.2 Subsector Shares of Agricultural Value Added, Thailand, 1975­2000 259 7.3 Estimates of Transmission Elasticities from Wholesale to Farm Prices, Thailand 272 7.4 NRAs for Covered Agricultural Products, Thailand, 1970­2004 274 7.5 NRAs in Agriculture Relative to Nonagricultural Industries, Thailand, 1970­2004 276 8.1 Structure of the Agricultural Economy, Vietnam, 1986­2004 285 8.2 Share of Planted Area by Crop, Vietnam, 1990­2004 286 8.3 Composition of Agricultural Exports by Value, Vietnam, 1990­2004 288 8.4 NRAs for Covered Agricultural Products, Vietnam, 1986­2004 296 8.5 NRAs in Agriculture Relative to Nonagricultural Industries, Vietnam, 1986­2004 297 9.1 Tariff Rates on Imports, Bangladesh, 1991­2003 313 9.2 Total Taxes on Agricultural Commodity Imports, Bangladesh, 1992 and 2002 314 9.3 Production and Imports, Rice and Wheat, Bangladesh, 1973­2004 316 9.4 NRAs for Covered Agricultural Products, Bangladesh, 1974­2004 321 9.5 NRAs in Agriculture Relative to Nonagricultural Industries, Bangladesh, 1974­2004 327 10.1 Sectoral Share of GDP and Employment, India, 1950­2005 341 10.2 Profile of Agricultural and Manufacturing Imports and Exports, India, 1960­2005 342 10.3 Structure of the Agricultural Economy, India, 1964­2004 357 10.4 Structure of Fertilizer and Electricity Subsidies, Key Crops, India, 2004 359 Contents xv 10.5 NRAs for Covered Agricultural Products, India, 1965­2004 361 10.6 NRAs in Agriculture Relative to Nonagricultural Industries, India, 1965­2004 362 10.7 Self-Sufficiency Ratios, Selected Food Products, India, 1961­2004 365 11.1 Agriculture in the Economy, Pakistan, 1965­2004 380 11.2 Structure of the Agricultural Economy, Major Crops, Pakistan, 2004­06 383 11.3 Production and Trade, Wheat, Pakistan, 1972­2005 387 11.4 NRAs for Covered Agricultural Products, Pakistan, 1962­2005 389 11.5 Production and Trade, Cotton, Pakistan, 1960­2005 391 11.6 NRAs in Agriculture Relative to Nonagricultural Industries, Pakistan, 1962­2005 400 12.1 Agriculture in the Economy, Sri Lanka, 1950­2005 413 12.2 NRAs for Covered Agricultural Products, Sri Lanka, 1955­2004 421 12.3 NRAs in Agriculture Relative to Nonagricultural Industries, Sri Lanka, 1955­2004 422 B.1 Annual Distortion Estimates, Republic of Korea, 1955­2004 475 B.2 Annual Distortion Estimates, Taiwan, China, 1955­2002 481 B.3 Annual Distortion Estimates, China, 1955­2005 487 B.4 Annual Distortion Estimates, Indonesia, 1970­2004 490 B.5 Annual Distortion Estimates, Malaysia, 1960­2004 496 B.6 Annual Distortion Estimates, the Philippines, 1962­2004 502 B.7 Annual Distortion Estimates, Thailand, 1970­2004 506 B.8 Annual Distortion Estimates, Vietnam, 1986­2005 510 B.9 Annual Distortion Estimates, Bangladesh, 1974­2004 513 B.10 Annual Distortion Estimates, India, 1965­2004 517 B.11 Annual Distortion Estimates, Pakistan, 1962­2005 523 B.12 Annual Distortion Estimates, Sri Lanka, 1955­2004 527 B.13 Annual Distortion Estimates, Asian Focus Economies, 1955­2005 533 B.14 Gross Subsidy Equivalents of Assistance to Farmers, Asian Focus Economies, 1955­2005 543 B.15 Share of the Regional Value of Agricultural Production, Asian Focus Economies, 1955­2004 546 B.16 Summary of NRA Statistics, Asian Focus Economies 549 B.17 Summary of NRA Statistics, by Major Product, Asian Focus Economies, 2000­04 550 B.18 Share of the Global Value of Production and Consumption, Key Covered Products, Asian Focus Economies, 2000­04 552 B.19 Share of the Global Value of Exports and Imports, Key Covered Products, Asian Focus Economies, 2000­03 555 B.20 Share of Production for Export, Consumption of Imports, and Domestic Production, Key Covered Products, Asian Focus Economies, 2000­03 558 FOREWORD Three-quarters of the world's poorest households depend on farming for their livelihoods, and the majority live in Asia where 81 percent of the poor (more than 900 million people earn less than US$1.25/day) are engaged directly or indi- rectly in agriculture. During the 1960s and 1970s, many developing countries had in place pro-urban and anti-agricultural policies, while many high-income countries restricted agricultural imports and subsidized their farmers. Although progress has been made over the past two decades to reduce those policy biases, particularly the anti-agricultural bias in Asia, the extent of reform has not been systematically quantified. Nor has it been clear how many trade- and welfare- reducing price distortions remain in Asian agriculture, both within and between countries, and whether later developing countries have followed Japan and the Republic of Korea in replacing past anti-agricultural policies with not a neutral regime but a pro-agricultural set of policies--which could be just as wasteful of national resources. To help fill this lacuna, the World Bank launched a major research project in 2006 aimed at quantifying the changing extent of distortions to agricultural incen- tives over recent decades. This volume is one of a series of four regional books that summarizes the findings. By including all the large Asian economies as case studies, no less than 95 percent of Asian GDP and agricultural output is covered. These estimates are used to help address questions such as the following: Where is there still a policy bias against agricultural production? To what extent has there been overshooting in the sense that some developing-country farmers are now being protected from import competition? What are the political and economic forces behind the more successful reformers, and how do these forces compare with those in less successful countries where major distortions in agri- cultural incentives remain? How important have domestic political forces been in xvii xviii Foreword bringing about reform, compared with international forces? What explains the cross-commodity pattern of distortions within the agricultural sector of each country? What policy lessons and trade implications can be drawn from these dif- fering experiences with a view to ensuring better growth-enhancing and poverty- reducing outcomes in other still-distorted developing countries during their reforms in the future? In Asia more than anywhere else, the reforms have been truly transformational. The world's most populous nations, China and India, have been among the most ambitious in raising incentives for farmers, albeit from a very low base in each case. Vietnam also has undertaken major and rapid reforms, while in other South- east Asian economies the reforms have been more gradual (or nonexistent in the cases of Myanmar and the Democratic People's Republic of Korea, but they were not able to be included as case studies because of lack of access to data). Mean- while, the Republic of Korea has moved from taxing agriculture relative to other tradable sectors in the 1950s to increasingly protecting it beginning in the 1970s. This development has raised concerns that other emerging economies may follow suit and pursue the same agricultural protection growth path of more-advanced economies. The new empirical indicators summarized in these case studies provide a strong evidence-based foundation for assessing the successes and failures of poli- cies of the past and for evaluating policy options for the years ahead. The analyti- cal narratives reveal that the reforms to agricultural price and trade policies were sometimes undertaken unilaterally. In other cases, they were also partly in response to international pressures such as the Uruguay Round (for example, the Republic of Korea), commitments required for accession to the World Trade Organization (WTO) (for example, China), and structural adjustment loan con- ditionality by international financial institutions (for example, the Philippines in the 1980s). The study is timely because the WTO is in the midst of the Doha round of mul- tilateral trade negotiations, and agricultural policy reform is one of the most con- tentious issues in those talks. Hopefully China and South and Southeast Asian countries will not make use of the legal wiggle room they have allowed themselves in their WTO bindings and follow Japan and the Republic of Korea into high agri- cultural protection. It might be argued, on one hand, that a laissez-faire strategy could increase rural-urban inequality and poverty and thereby generate social unrest. On the other hand, policies that lead to high prices for staple foods, in par- ticular, involve potentially serious risks for the urban and rural poor who are net buyers of food in developing countries. Available evidence suggests that problems of rural-urban poverty gaps have been alleviated in parts of Asia by some of the more-mobile members of farm households finding full- or part-time work off the Foreword xix farm and repatriating part of their higher earnings back to those remaining in farm households. Efficient ways of assisting any left-behind groups of poor (non- farm as well as farm) households include public investment measures that have high social payoffs, such as in basic education and health and in rural infrastruc- ture, as well as in agricultural research and development. As argued in the World Bank's World Development Report 2008, the latter also provide more sustainable and more equitable ways of securing domestic food supplies than artificially propping up prices. Justin Yifu Lin Senior Vice President and Chief Economist The World Bank ACKNOWLEDGMENTS This book provides an overview of the evolution of the distortions to agricultural incentives caused by price and trade policies in the World Bank­defined regions of East Asia and South Asia. The volume includes an introduction and summary chapter and commissioned studies of three Northeast Asian, five Southeast Asian, and four South Asian economies. The chapters are followed by two appendixes. The first appendix describes the methodology we have used to measure the nom- inal and relative rates of assistance for farmers and the taxes and subsidies on food consumption. The second appendix provides summaries of our annual estimates of these rates of assistance across the focus economies. Together, the 12 economies we study account for no less than 95 percent of the region's agricultural value added, farm households, total population, and total gross domestic product. To the authors of the case studies, who are listed on the following pages, we are extremely grateful for the dedicated way in which they have delivered far more than we could have reasonably expected. We are particularly grateful to Yujiro Hayami for his insights and advice during this project and his influential, related work over several decades. Staff at the World Bank's East Asia and Pacific Depart- ment and South Asia Department have provided generous and insightful advice and assistance throughout the project. This has included participation in two Bank-wide seminars that provided helpful suggestions on the draft studies. We offer thanks likewise to the World Bank directors in the focus economies, who examined and cleared the working paper versions of each chapter. We have simi- larly benefited from the feedback provided by many participants at workshops and conferences in which drafts have been presented over the past year or so. Johanna Croser, Francesca de Nicola, Esteban Jara, Marianne Kurzweil, Signe Nelgen, Damiano Sandri, and Ernesto Valenzuela have generously assisted in xxi xxii Acknowledgments compiling material for the opening overview chapter, and Johanna Croser and Marie Damania assisted in copyediting the case study chapters. We wish to extend our thanks to the Organisation for Economic Co-operation and Development and the International Food Policy Research Institute for sharing methodological insights, and also to the members of the project's Senior Advisory Board, who have provided sage advice and much encouragement throughout the planning and implementation stages. The board is comprised of Yujiro Hayami, Bernard Hoekman, Anne Krueger, John Nash, Johan Swinnen, Stefan Tangermann, Alberto Valdés, Alan Winters, and, until his untimely death in 2008, Bruce Gardner. Our thanks go also to the Development Research Group and to the Trust Funds of the governments of Ireland, Japan, the Netherlands, and the United Kingdom for financial assistance. This support has made it possible for this set of economies to be included as part of a wider study that also encompasses more than 30 other developing countries, 18 economies in transition from central planning, and 20 high-income countries. Three companion volumes examine case studies of other emerging economies in a similar way and for a similar time period (back to the mid-1950s or early 1960s, except for the transition economies). Also published by the World Bank in 2008 or 2009, they cover Africa (coedited by Kym Anderson and Will Masters), Latin America (coedited by Kym Anderson and Alberto Valdés), and Europe's transition economies (coedited by Kym Anderson and Johan Swinnen). A global overview volume edited by Kym Anderson will be pub- lished in 2009. Kym Anderson and Will Martin November 2008 CONTRIBUTORS Nazneen Ahmed is a research fellow at the Bangladesh Institute of Development Studies, Ministry of Planning in Dhaka, Bangladesh. Kym Anderson is George Gollin Professor of Economics at the University of Adelaide and a fellow of the Center for Economic Policy Research, London. During 2004­07, he was on an extended sabbatical as lead economist (trade policy) in the Development Research Group of the World Bank in Washington, DC. Prema-Chandra Athukorala is professor of economics in the Research School of Pacific and Asian Studies at the Australian National University in Canberra. Arsenio M. Balisacan is professor of agricultural economics and director of the Southeast Asian Regional Center for Graduate Study and Research in Agriculture in Los Baños, Laguna, the Philippines. Zaid Bakht is research director at the Bangladesh Institute of Development Stud- ies, Ministry of Planning in Dhaka, Bangladesh. Jayatilleke Bandara is associate professor of economics at Griffith University in Brisbane, Australia. Johanna Croser has been a consultant with this project and is a PhD student in the Department of Economics of the University of British Columbia in Vancouver, Canada. Cristina David is a senior research fellow at the Philippine Institute for Develop- ment Studies in Makati City, the Philippines. xxiii xxiv Contributors Paul A. Dorosh is a senior rural development economist at the Agriculture and Rural Development Department (South Asia Agriculture and Rural Development Group) at the World Bank in Washington, DC. George Fane is an adjunct professor of economics in the Research School of Pacific and Asian Studies, Australian National University in Canberra. Ashok Gulati is the Asian Director for the International Food Policy Research Institute in New Delhi. Previously, he headed the institute's Markets, Trade, and Institutions Division. Kanupriya Gupta is a senior research analyst with the International Food Policy Research Institute in New Delhi. Yujiro Hayami is chairman of the graduate faculty of the Foundation for Advanced Studies on International Development. He is also visiting professor in the National Graduate Institute of Policy Studies, Tokyo. Masayoshi Honma is professor of agricultural and resource economics at the University of Tokyo. He is also a member of the Board of Trustees of the Interna- tional Food Policy Research Institute in Washington, DC. Jikun Huang is director and professor, Center for Chinese Agricultural Policy, Chinese Academy of Sciences in Beijing. Pham Lan Huong is a researcher with the Central Institute of Economic Manage- ment in Hanoi, Vietnam. Ponciano Intal is professor and director, Angelo King Institute for Economic and Business Studies, De La Salle University in Manila. Sisira Jayasuriya was director, Asian Economics Center and associate professor in the Department of Economics, University of Melbourne at the time of the study and is now professor of economics at La Trobe University (Bundoora) in Victoria, Australia. Archanun Kohpaiboon is lecturer in the economics department of Thammasat University in Bangkok. Marianne Kurzweil is a young professional at the African Development Bank in Tunis. During 2006­07, she was consultant with this project at the Development Research Group at the World Bank in Washington, DC. Contributors xxv Yu Liu is researcher at the Center for Chinese Agricultural Policy, Chinese Acad- emy of Sciences in Beijing. Wai-Heng Loke is lecturer at the Department of Economics, Faculty of Economics and Administration at the University of Malaya in Kuala Lumpur, Malaysia. Will Martin is lead economist in the Development Research Group at the World Bank in Washington, DC. He specializes in trade and agricultural policy issues globally, especially in Asia. Signe Nelgen has been a consultant with this project and is a PhD student in the School of Economics of the University of Adelaide in Australia. Garry Pursell is visiting fellow at the Australia South Asia Research Center, Australian National University in Canberra. Previously, he was with the South Asia Department of the World Bank. Scott Rozelle holds the Helen Farnsworth Endowed Professorship and is senior fellow and professor at the Freeman Spogli Institute for International Studies at Stanford University in Stanford, California. Abdul Salam is professor of economics at the Federal Urdu University and is for- mer chairman of the Agricultural Prices Commission in Islamabad, Pakistan. Damiano Sandri is a PhD candidate in economics at Johns Hopkins University in Baltimore, Maryland. During 2006­07, he was a consultant with this project at the Development Research Group at the World Bank in Washington, DC. Quazi Shahabuddin is director general of the Bangladesh Institute of Develop- ment Studies, Ministry of Planning in Dhaka, Bangladesh. Vo Tri Thanh is a researcher with the Central Institute of Economic Management in Hanoi, Vietnam. Ernesto Valenzuela is lecturer in economics and research fellow at the University of Adelaide in Australia. During 2005­07, he was consultant at the Development Research Group of the World Bank in Washington, DC. Peter Warr is the John Crawford Professor of Agricultural Economics and found- ing Director of the Poverty Research Centre in the Division of Economics, Research School of Pacific and Asian Studies at the Australian National University in Canberra. ABBREVIATIONS CTE consumer tax equivalent GDP gross domestic product GVA gross value added IMF International Monetary Fund NRA nominal rate of assistance NRP nominal rate of protection NTB nontariff barrier OECD Organisation for Economic Co-operation and Development RRA relative rate of assistance TBI trade bias index WTO World Trade Organization Note: All dollar amounts are U.S. dollars (US$) unless otherwise indicated. xxvii The Focus Economies of Asia REP. OF CHINA KOREA PAKISTAN TAIWAN, INDIA CHINA THAILAND BANGLADESH PHILIPPINES VIETNAM SRI LANKA 0 500 1000 Kilometers MALAYSIA 0 500 1000 Miles This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information INDONESIA shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any IBRD 36674 endorsement or acceptance of such boundaries. DECEMBER 2008 Part I II INTRODUCTION 1 INTRODUCTION AND SUMMARY Kym Anderson and Will Martin Farm earnings in developing countries have often been depressed by a pro-urban, antiagricultural bias in government policies. Progress has been made since the 1980s in reducing the policy bias in many countries, however. In some cases, the changes have been modest and intermittent, but, in China and, to a lesser extent, India, they have been transformational. Many trade-reducing price distortions nonetheless remain within the agricultural sector in low- and middle-income economies, including in Asia. This is important for the majority of households in the world because 45 percent of the global workforce is employed in agriculture, and 75 percent of the world's poorest households depend directly or indirectly on farming for livelihoods. It is even more important in Asia's developing economies, where 60 percent of the workforce and 81 percent of the poor (625 million people each earning less than US$1 a day) are engaged in agriculture (World Bank 2007; Chen and Ravallion 2007). This study is part of a global research project seeking to understand the chang- ing scope and impact of the policy bias against agriculture and the reasons behind agricultural policy reforms in Africa, Europe's transition economies, Latin Amer- ica and the Caribbean, and Asia.1 One purpose of the project is to obtain quanti- tative indicators of the effects of recent policy interventions. A second objective is to gain a deeper understanding of the political economy of trends in the distor- tions in agricultural incentives in various national settings. The third goal is to use this deeper understanding to explore the prospects for reducing the distortions in agricultural incentives and discover the likely implications for agricultural com- petitiveness, equality, and poverty reduction in many countries, large and small. 3 4 Distortions to Agricultural Incentives in Asia The compilation of new annual time series estimates of the protection and tax- ation of farmers over the past half century is a core component of the first stage of our research project. These estimates are used to help address questions such as the following: Where is the policy bias against agricultural production? To what extent has there been overshooting in that some food producers are now being protected from import competition in developing countries in much the same way they were protected in Europe and Japan during earlier periods of industrial- ization? What are the political economy forces behind successful reforms in some countries? How do they compare with the forces in other countries where reforms are less successful and major distortions in agricultural incentives remain? Over the past two decades, how important have domestic political forces been in gener- ating reform relative to forces operating in earlier decades or international forces, such as loan conditionality, rounds of multilateral trade negotiations in the World Trade Organization (WTO), regional integration agreements, accession to the WTO, and the globalization of supermarkets and other firms along the value chain? What has caused the patterns of distortion within the agricultural sectors of individual countries? What policy lessons and trade implications may be drawn from these differing experiences so that we may seek to ensure growth-enhancing and poverty-reducing outcomes, including less overshooting and fewer protec- tionist regimes, during future reforms in still-distorted developing economies in Asia and elsewhere? Our study is timely for at least four reasons. First, the WTO is in the midst of the Doha Round of multilateral trade negotiations, and agricultural policy reform is one of the most contentious issues in these talks. Second, countries are also seeking to position themselves favorably in preferential trade negotiations in the wake of other forces in globalization, including revolutions in information, communica- tions, agricultural biotechnology, and supermarketing. Third, poorer countries and their development partners are striving to achieve the United Nations Millennium Development Goals by 2015, including the prime goals of alleviating hunger and reducing poverty. Fourth, the outputs of our study are timely also because world food prices spiked at high levels in 2008, and governments in some developing countries, in their panic to deal with the inevitable protests from consumers, have reacted in ways that are not optimal. Spikes have occurred in the past, most notably in 1973­74, and lessons on appropriate policy responses may be drawn from the experiences then. The empirical estimates reported in our study reveal that govern- ments in Asia have differed in their responses to such shocks, although this is less the case of rice, the region's main staple. This study on Asia is based on a sample of 12 developing economies. We exclude Japan, which has been a high-income country throughout our review period and, so, is analyzed separately as part of the high-income group in the Introduction and Summary 5 project's global overview volume. In Northeast Asia, we include China, the Republic of Korea (hereafter referred to as Korea), and Taiwan, China. In Southeast Asia, we include the five large economies of Indonesia, Malaysia, the Philippines, Thailand, and Vietnam, and, in South Asia, we include the four largest economies: Bangladesh, India, Pakistan, and Sri Lanka. In 2000­04, these economies (all of them now WTO members) accounted for more than 95 per- cent of Asia's agricultural value added, total farm households, total population, and total gross domestic product (GDP).2 The distortion estimates are provided for as many years as data permit over the past five decades (an average of 42 years), and they are presented for an average of eight crop and livestock prod- ucts per economy, which, in aggregate, amounts to about 70 percent of the gross value of agricultural production in each of these economies. The time series and country coverage in our study greatly exceed the data and country coverage of ear- lier studies, including Anderson and Hayami (1986); Krueger, Schiff, and Valdés (1991); Orden et al. (2007); and the Organisation for Economic Co-operation and Development (OECD 2007). The product coverage is broader in each of our case studies than in all earlier case studies other than the study by the OECD (2007).3 The key characteristics of these economies--accounting in 2000­04 for only 10 percent of worldwide GDP, but 37 percent of global agricultural value added, 51 percent of the world's population, and 73 percent of the world's farmers--are shown in table 1.1. The table reveals the considerable diversity within the region in development, relative resource endowments, comparative advantage, trade spe- cialization, and the incidence of poverty and income inequality. These economies thus provide a rich sample for comparative study. Per capita incomes in Bangladesh, India, and Vietnam are barely 8 percent of the world average. In Indonesia, the Philippines, and Sri Lanka, they are around 16 percent. In China, they are over 25 percent. In Thailand, they are more than 30 percent. In Malaysia, they are about 75 percent of the world aver- age. Korea and Taiwan, China appear exceptional in that average per capita incomes are currently twice the global average; however, in the 1950s, at the start of the period of our study, these two economies were among the poorest in the world. Korea and Taiwan, China are also exceptional in the per capita endowment of agricultural land; they have only around 5 percent of the world average endow- ment ratio. Bangladesh has a little more, followed by Sri Lanka and the Philip- pines. Even India, Indonesia, and Pakistan have only about 25 percent of the global average endowment, while Malaysia and Thailand have about 40 percent, and China, over 50 percent.4 Thus, these Asian economies are not relatively well endowed with cropland or pastureland; on a per capita basis, the region has only Table 1.1. Key Economic and Trade Indicators, Asian Focus Economies, 2000­04 6 Agricul- Share of world, % Index, world 100 tural Gini index, trade per capita Agricul- Agricul- Agricul- speciali- incomed Economy or Popula- Total tural tural GDP per tural land zation Poverty subregion tion GDP GDP workers capita per capita RCAa indexb incidencec 1984 2004 East Asia 29.09 8.38 24.76 47.1 29 45 75 0.12 9 24 37 China 20.60 4.33 16.62 38.4 21 54 58 0.05 10 20 36 Indonesia 3.41 0.59 2.62 3.8 17 27 173 0.08 4 30 35 Korea, Rep. of 0.77 1.62 1.69 0.2 212 5 26 0.78 0 -- -- Malaysia 0.39 0.28 0.73 0.1 74 41 107 0.18 0 49 49 Philippines 1.27 0.22 0.91 1.0 18 19 67 0.10 13 41 44 Taiwan, China 0.36 0.84 0.45 0.1 232 5 28 0.72 0 -- -- Thailand 1.01 0.38 1.05 1.5 38 39 204 0.38 1 45 42 Vietnam 1.29 0.11 0.69 2.1 8 14 301 0.61 1 36 37 South Asia 21.67 1.99 11.90 25.3 9 20 145 0.07 31 31 35 Bangladesh 2.16 0.14 0.90 2.9 7 8 93 0.69 35 26 33 India 16.87 1.57 9.32 20.2 9 22 143 0.24 36 31 33 Pakistan 2.33 0.23 1.43 1.9 10 23 137 -- 17 33 31 Sri Lanka 0.31 0.05 0.24 0.3 16 15 254 0.45 6 32 40 Total 50.76 10.37 36.65 72.5 20 34 80 0.15 19 27 36 Sources: Sandri, Valenzuela, and Anderson 2007; World Development Indicators Database 2008; PovcalNet 2008. Note: -- no data are available. a. The index of revealed comparative advantage (RCA) for agriculture and processed foods is the share of agriculture and processed food in national exports as a ratio of such products in worldwide exports. b. The index of primary agricultural trade specialization is defined as net exports, divided by the sum of the exports and imports of agricultural and processed food products (the world average 0.0). c. The percentage of the population living on less than US$1 a day. d. Poverty incidence and the 2004 Gini index are for the most recent year available between 2000 and 2004. The 1984 Gini index is for the available year nearest 1984 (PovcalNet). The weighted averages for the focus economies use population as the basis for weights. Introduction and Summary 7 34 percent of the global average. This might suggest that the comparative advan- tage of the Asian economies in agricultural goods is low, were it not for the varia- tions in these economies in the level of industrial development, the quality of land and water, and the related institutional arrangements and entitlements. As a result, the strengths of these economies in agricultural competitiveness are diverse. The differences are reflected in the index of revealed comparative advan- tage and the agricultural trade specialization index (table 1.1). A majority of our focus economies have an index of revealed comparative advantage well above 100, indicating the extent to which the share of agricultural and food products in an economy's merchandise trade exceeds the global average share of these products. For Korea and Taiwan, China, the index is below 30, and, for China and the Philip- pines, it is around 60. The index of agricultural trade specialization measures net exports as a ratio of exports, plus imports of farm products, and, so, it is bounded between 1 and 1. It is positive for half of our focus economies, but is 0.7 for Bangladesh, Korea, and Taiwan, China. Income inequality has risen slightly over the past two decades, but is still low throughout much of the region relative to the rest of the world. In 2004, the Gini coefficient was between 0.40 and 0.49 in Malaysia, the Philippines, Sri Lanka, and Thailand and averaged between 0.31 and 0.37 in the rest of the region. The regional average of 0.36 contrasts with, for example, the average of 0.52 in Latin America. Likewise, the Gini coefficient for land distribution is relatively low in Asia, at only 0.41 for China and Pakistan and below 0.50 also in Bangladesh; Indonesia; Korea; Taiwan, China; and Thailand. Even in India, the coefficient for land distribution is only 0.58, and it is 0.50 in Vietnam. However, these coefficients imply that the distri- bution of land is more equal in Asia than it is in Latin America, where the Gini coef- ficient for land distribution is above 0.70 for major countries such as Argentina and Brazil and possibly for the region as a whole (World Bank 2007). A significant pro- portion of the rural population is landless in South Asia, however; so 31 percent of the population of South Asia was still living on less than US$1 a day in 2004 com- pared with only 9 percent in East Asia (table 1.1). The extent of the decline in poverty in Asia has been unprecedented. The num- ber of people living on less than US$1 a day has been reduced by half since 1981 (in 1993 purchasing power parity dollars). Most of the decline has occurred in East Asia, especially China. In East Asia, the poverty rate declined from 58 percent to less than 10 percent of the population; but, even in South Asia, the proportion has fallen from 50 to around 30 percent (table 1.2). During the 10 years to 2002, no less than 75 percent of the decline in the share of the poor among the popula- tion in Asia occurred in rural areas, and another 15 percent of the decline was gen- erated by a movement out of poverty among rural people who had migrated because of better opportunities in urban areas (Chen and Ravallion 2007). 8 Distortions to Agricultural Incentives in Asia Table 1.2. Poverty Levelsa in Asia, 1981­2004 Economy, indicator 1981 1987 1993 1999 2004 Poor people, millions Asia 1,251 900 857 740 615 China 634 310 334 223 128 Other East Asia 162 119 86 53 41 India 364 369 376 376 371 Other South Asia 91 102 61 87 75 Population share, % East Asia 58 28 25 15 9 South Asia 50 45 37 35 31 Source: Chen and Ravallion 2007. a. People living on less than US$1 a day at 1993 international purchasing power parity. Policy developments have made nontrivial contributions to the growth, struc- tural change, and poverty reduction observed in Asia over the past five decades. The transformational shift from central planning and state-owned enterprises to greater dependence on markets and private entrepreneurship has had a particularly dra- matic effect in China and Vietnam since the 1980s, but India has also benefited from similar reforms beginning in the early 1990s. Also important has been the abandon- ment in market economies of import-substituting industrialization in favor of export-oriented development strategies, beginning in Taiwan, China around 1960 and followed by Korea, then by several Southeast Asian economies, and now also by economies in South Asia. Agricultural policies have not been the only or even the main target of these reforms, but they have been an integral part of the process. We begin with a brief summary of economic growth and the structural changes in the region since the 1950s. We also examine the agricultural and other economic policies that have affected agriculture before and after the various reforms and, in several cases, after fundamental regime changes during the past half century.5 We then introduce the methodology used by the authors of the indi- vidual country studies to estimate the nominal rates of assistance (NRAs) and rel- ative rates of assistance (RRAs) for farmers delivered by national farm and non- farm policies over the past several decades (depending on data availability), as well as the impact of these policies on the consumer prices of farm products. Farmer assistance and consumer taxation will be negative during periods of antiagricul- tural, pro-urban bias in a policy regime. We subsequently provide a synopsis of the empirical results detailed in the country studies in this volume, though we do not attempt to survey the myriad policy changes discussed in detail in the follow- ing chapters. The final sections of this chapter summarize what we have learned Introduction and Summary 9 and draw out the implications of the findings for poverty, inequality, and the possible future direction of policies affecting agricultural incentives in Asia. Growth and Structural Change The most striking economic characteristic of the developing economies in Asia, particularly East Asia, is the rates of economic growth and industrial development in Korea; Taiwan, China; and elsewhere over the past three decades or more (Anderson 2008). The recent report of the Commission on Growth and Develop- ment (2008) has noted that 13 of the world's economies have had sustained growth in real per capita income of more than 7 percent for at least 25 consecutive years since World War II, and nine of these economies are in East Asia.6 Between 1980 and 2004, per capita GDP grew 6.3 percent per year in East Asia and 3.4 per- cent per year in South Asia, while the global average was only 1.4 percent. Asia's industrial growth during this period was 8.6 percent per year. This compares with the world average of 2.5 percent. Even the agricultural growth rate was more than half again as high in Asia relative to the world average (3.1 percent and 2.0 percent per year, respectively; see table 1.3). As a consequence of this growth performance, per capita incomes in some Asian economies have been converging rapidly, albeit from a low base, toward incomes in rich countries, while other developing and transition economies have, on average, been slipping away from the performance of rich countries such as the United States (figure 1.1). A key driver of the rapid growth and industrialization in Asia has been the decision by many governments in the region to open up economies and abandon an import-substituting development strategy in favor of an export-oriented approach. This shift occurred at different times in our focus economies, beginning with Korea and Taiwan, China in the 1960s. China joined the group in the late 1970s, Vietnam in the mid-1980s, and India haltingly in the early 1980s and more concertedly in 1991. As a result, export volumes grew at double-digit rates (last column of table 1.3). The share of exports in GDP rose steadily in the region, more than doubling in the 30 years to 2004 (table 1.4). East Asia's share in world- wide exports of nonfood manufactures has quadrupled since 1990, thanks espe- cially to industrialization in China. China accounted for 11 percent of the world's manufacturing exports in 2006, compared with less than 1 percent in 1990: a 20- fold increase in current U.S. dollar terms. Our other focus economies experienced an average fivefold increase, and all the economies have contributed to the region's growing share in global manufacturing exports since 1990 (table 1.5).7 Along with the export-led growth has come a dramatic restructuring of Asia's economies that has involved a shift from agriculture toward manufacturing and service activities. In East Asia, the share of the agricultural sector in GDP is now 10 Table 1.3. Real Growth in GDP and Exports, Asian Focus Economies, 1980­2004 (at constant 2000 prices, trend-based, % per year) GDP (1980­2004) Export volume, Economy Agriculture Industry Services Total Per capita 1985­95 East Asia 3.1 9.0 7.9 7.6 6.3 13.7 China 4.4 12.1 11.3 9.9 8.6 15.1 Indonesia 2.9 6.6 5.3 5.4 3.7 10.4 Korea, Rep. of 1.3 8.2 7.2 7.1 6.1 10.6 Malaysia 1.7 7.8 6.9 6.6 3.9 10.3 Philippines 1.7 2.0 3.5 2.7 0.4 12.8 Taiwan, China 0.5 5.3 8.3 6.7 5.6 17.0 Thailand 2.4 8.5 5.8 6.3 4.9 17.3 Vietnam 3.9 9.7 7.5 7.0 5.1 -- South Asia 3.0 6.2 6.4 5.4 3.4 -- Bangladesh 2.7 6.6 4.4 4.4 2.1 13.4 India 3.0 6.3 7.0 5.7 3.7 -- Pakistan 4.0 5.5 4.8 4.7 2.1 9.8 Sri Lanka 1.8 5.6 5.1 4.5 3.3 6.3 Total 3.1 8.6 7.5 7.1 5.5 -- World 2.0 2.5 3.2 3.0 1.4 -- Sources: Sandri, Valenzuela, and Anderson 2007; World Development Indicators Database 2008. Note: -- no data are available. Figure 1.1. Index of Real Per Capita GDP, Asia Relative to the United States, 1950­2006 a. Asian focus economies 70 60 100 50 40 States 30 United 20 index, 10 0 ­54 1950 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­06 year China Korea, Rep. of and Southeast Asia South Asia Taiwan, China b. Asia and other developing and transition regions 40 100 30 States 20 United 10 index, 0 ­54 1950 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 2005­06 year Africa Latin America Europe's transition economies Asia, excluding Japan Sources: Author calculations; Maddison 2003; World Development Indicators Database 2008. Note: The charts are based on 1990 Geary-Khamis international dollars up to 2001, taken from Maddison (2003), and updated using real GDP per capita growth data from the World Development Indicators Database. The economies and regions are indicated relative to the United States, which is set as the numeraire at 100. Southeast Asia is Indonesia, Malaysia, the Philippines, Thailand, and Vietnam. South Asia is Bangladesh, India, Pakistan, and Sri Lanka. 11 12 Table 1.4. Exports of Goods and Services as a Share of GDP, Asian Focus Economies, 1965­2004 (percent) Economy 1965­69a 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 East Asiab 8 19 22 25 28 34 39 China 3 6 11 14 22 21 28 Indonesia 10 23 25 23 26 32 35 Korea, Rep. of 13 32 34 35 27 33 39 Malaysia 37 49 53 63 82 103 117 Philippines 11 20 21 25 28 47 53 Taiwan, China 22 49 53 54 45 48 59 Thailand 18 21 23 30 38 49 68 Vietnam -- -- -- -- -- 44 55 South Asia -- 7 7 7 11 12 14 Bangladesh -- 5 5 6 8 13 15 India 3 6 6 6 9 11 13 Pakistan -- 12 12 14 17 16 16 Sri Lanka 19 24 29 26 32 36 37 Totala -- 16 19 22 25 30 34 Sources: Sandri, Valenzuela, and Anderson 2007; World Development Indicators Database 2008. Note: -- no data are available. a. Only merchandise exports in 1965­69, except for Taiwan, China. In 1960­64, the shares of Korea and Taiwan, China were 6 and 15 percent, respectively. In 1955­59, the share of Taiwan, China was 10 percent. b. Ignores Vietnam in 1980­94, when the weight of Vietnam in Asian GDP was less than 1 percent. Introduction and Summary 13 Table 1.5. Share of Nonfood Manufactures in World Exports, Asian Focus Economies, 1990­2006 (percent) 2006 value as Economy 1990­94 1995­99 2000­04 2006 % of 1990 East Asia 5.1 10.3 15.6 20.4 790 China 1.0 2.4 6.1 10.8 2,020 Indonesia 0.1 0.5 0.6 0.5 490 Korea, Rep. of 1.7 2.6 3.1 3.5 480 Malaysia 0.2 1.0 1.5 1.4 750 Philippines 0.1 0.2 0.6 0.5 730 Taiwan, China 1.8 2.7 2.5 2.3 310 Thailand 0.2 0.8 1.0 1.2 670 Vietnam 0.0 0.1 0.2 0.2 -- South Asia 0.6 0.9 1.1 1.4 470 Bangladesh 0.0 0.1 0.1 0.1 800 India 0.4 0.5 0.7 1.0 550 Pakistan 0.1 0.2 0.2 0.2 310 Sri Lanka 0.0 0.1 0.1 0.1 480 Total 5.7 11.2 16.7 21.8 760 Sources: Sandri, Valenzuela, and Anderson 2007; WTO 2007; World Development Indicators Database 2008. Note: -- no data are available. less than 30 percent of the sector's share in the late 1960s. In slower-growing South Asia, the share has fallen by less than 50 percent over the period. The biggest changes have occurred in China and Indonesia, where agriculture's share in GDP dropped from more than 40 and 50 percent, respectively, in the 1960s to 13 per- cent in 2005. Bangladesh has also been transformed remarkably: agriculture's share was 54 percent in 1965­69, and all other nonservice industries accounted for only 9 percent; now, industry's share is higher than the share of agriculture (27 versus 20 percent in 2005). Pakistan, the Philippines, and Sri Lanka are the focus economies that have been growing the most slowly, and they have also been the slowest in moving away from agriculture since the 1960s. At the other extreme are Korea and Taiwan, China; only 2 or 3 percent of the GDP of these economies is now derived from farming. For Asia as a whole, agriculture now accounts for only around 12 percent of GDP, down from about 36 percent in the late 1960s. Mean- while, the share of industry has risen from 27 to 38 percent, and the share of serv- ices from 35 to 49 percent (table 1.6). The share of overall employment accounted for by farming activities has fallen somewhat more slowly than agriculture's GDP share, according to statistics in the 14 Table 1.6. Sectoral Shares of GDP, Asian Focus Economies, 1965­2004 (percent) Agriculture Industry Services Economy 1965­69 1975­79 1985­89 2000­04 1965­69 1975­79 1985­89 2000­04 1965­69 1975­79 1985­89 2000­04 East Asiaa 34 26 19 10 29 40 40 42 34 32 41 48 China 39 31 27 14 35 47 44 46 26 22 30 41 Indonesia 49 29 23 16 16 35 36 45 35 36 41 40 Korea, Rep. of 30 21 10 4 22 30 37 35 48 49 53 61 Malaysia 29 26 20 9 27 37 39 49 44 37 42 42 Philippines 27 29 24 14 27 36 35 32 46 35 42 53 Taiwan, China 20 10 5 2 34 44 45 31 47 46 50 67 Thailand 30 25 16 10 24 29 34 43 46 46 50 48 Vietnam -- -- 41 23 -- -- 27 39 -- -- 32 38 South Asia 43 36 29 21 18 21 23 24 39 42 48 55 Bangladesh 54 55 31 22 9 14 21 25 36 31 48 52 India 44 36 29 21 19 22 24 24 38 43 47 55 Pakistan 35 29 24 22 19 21 21 22 46 50 55 56 Sri Lanka 29 28 24 17 21 26 24 24 51 46 51 59 Totala 36 28 22 12 27 36 35 38 35 34 43 49 Sources: Sandri, Valenzuela, and Anderson 2007; World Development Indicators Database 2008. Note: -- no data are available. a. Ignores Vietnam in 1965-69 and 1975­79, when the weight of Vietnam in Asian GDP was less than 1 percent. Introduction and Summary 15 Table 1.7. Agriculture's Share in Employment, Asian Focus Economies, 1965­2004 (percent) Economy 1965­69 1975­79 1985­89 2000­04 East Asia 77 72 68 60 China 79 75 73 66 Indonesia 69 60 56 47 Korea, Rep. of 53 41 24 9 Malaysia 57 45 31 18 Philippines 60 54 48 39 Taiwan, China 45 25 15 7 Thailand 81 74 66 55 Vietnam 79 74 72 67 South Asia 74 70 65 57 Bangladesh 85 76 67 54 India 73 70 66 59 Pakistan 65 64 55 46 Sri Lanka 55 53 49 45 Total 76 71 67 59 Sources: Sandri, Valenzuela, and Anderson 2007; FAOSTAT Database 2008. FAOSTAT Database of the Food and Agriculture Organization of the United Nations (which, because of definitional differences, is not always consistent with official national databases). The employment share remains at much higher levels than the GDP share, implying relatively low labor productivity on farms. The most rapid declines have occurred in Korea and Taiwan, China, where the employment share in agriculture has fallen from around 50 percent to less than 10 percent over the past 40 years. Malaysia, too, has experienced a major decline, from 57 to 18 percent of the workforce. Elsewhere in Asia, however, the share of employment remains large in farming (table 1.7). The share would be somewhat less in full-time equivalent terms if more careful account were taken of part-time off-farm activities (for example, see Otsuka and Yamano 2006), but the share nonetheless underscores the importance of the incentives faced by farmers in the well-being of the majority of Asia's households. The average share of agriculture in Asia's merchandise exports has declined even more dramatically than the GDP share over the past four decades, from 45 percent to only 7 percent. During this period, the share of nonprimary goods in exports has doubled, to 85 percent. Among our 12 focus economies, only the exports of such goods from natural resource-rich Indonesia represent less than 75 percent of the total exports of the economy (table 1.8). The declining relative 16 Table 1.8. Sectoral Shares of Merchandise Exports, Asian Focus Economies, 1965­2004 (percent) Agriculture and processed food Other primary goods Other goods Economy 1965­69 1975­79 1985­89 2000­04 1965­69 1975­79 1985­89 2000­04 1965­69 1975­79 1985­89 2000­04 East Asiaa 46 30 18 7 10 16 12 7 43 53 62 85 China 51 35 19 5 5 17 14 4 44 48 53 90 Indonesia 49 26 21 15 48 72 55 29 2 2 24 55 Korea, Rep. of 21 11 5 2 9 2 2 6 70 87 92 92 Malaysia 61 55 36 10 32 27 24 11 5 17 40 78 Philippines 78 55 27 6 16 17 11 3 7 18 32 83 Taiwan, China 39 13 6 3 2 2 2 3 59 85 92 94 Thailand 79 67 46 18 14 10 3 4 4 20 50 75 Vietnam -- -- -- 27 -- -- -- 23 -- -- -- 48 South Asia 42 40 26 13 9 8 7 8 49 52 66 78 Bangladesh -- 37 28 8 -- 1 1 0 -- 61 71 91 India 38 35 22 13 11 9 9 11 50 55 67 75 Pakistan 53 38 30 12 2 5 1 2 45 57 68 85 Sri Lanka 96 79 47 23 2 8 5 2 1 9 46 75 Totala 45 32 18 7 10 14 12 7 44 53 62 85 Sources: Sandri, Valenzuela, and Anderson 2007; World Development Indicators Database 2008. Note: -- no data are available. a. Ignores Vietnam in 1985­89, when the weight of Vietnam in Asian merchandise trade was less than 1 percent. Introduction and Summary 17 importance of farm exports has been much more rapid in Asia than in the rest of the world: the index of the revealed agricultural comparative advantage for Asia-- defined as the share of agriculture and processed food in national exports as a ratio of the share of such products in worldwide merchandise exports--has fallen since the 1980s by about two-thirds in East Asia and one-third in South Asia. The index of agricultural trade specialization (defined as net exports, divided by the sum of imports and exports of agricultural and processed food products) has also fallen. The latter index, by definition, ranges from 1 to 1. It has become increasingly less positive, or it has become negative in virtually all our Asian focus economies in recent decades (table 1.9). This apparent decline in agricultural comparative advantage is evident in the self-sufficiency data on primary farm products. Until 30 years ago, the region was almost exactly 100 percent self-sufficient in farm products; but, since then, the indicator has declined to less than 85 percent. The share of farm production that is exported has not changed much, averaging in the 4­6 percent range. However, there have been substantial changes in individual economies, including declines in Malaysia, the Philippines, Sri Lanka, and Taiwan, China and increases in China, Thailand, and Vietnam. In contrast, since the late 1970s, the share of imports in the domestic consumption of farm products has quadrupled, to around 20 per- cent (table 1.10). The growing dependence on imports of farm products in Asia has occurred despite reductions in the taxation of agricultural exports and increases in the incen- tives provided to farmers through government policy reforms (discussed below). These reforms have probably contributed to poverty reduction in Asia. Using a price indicator that is simpler than the one developed by us below, Ravallion and Chen (2007) show that the reduction in the antiagricultural bias in farm price policies has contributed significantly to poverty reduction in China. Rural growth is also a key contributor to the reduction in poverty in India (Ravallion and Datt 1996). One may revisit these and other, similar studies using the more comprehensive measures of the extent of changes in distortions to agricultural incentives summarized below. To gen- erate these measures, a common methodology has been adopted by the authors of the country case studies in this volume. A summary of the methodology follows, and additional details may be found in Anderson et al. (2008) and in appendix A. Methodology for Measuring Rates of Assistance or Taxation The NRA is defined as the percentage by which government policies have raised (or lowered if the NRA is less than 0) the gross returns to producers above (or below) the gross returns they would have received without government intervention. If a (Text continues on page 22.) 18 Distortions to Agricultural Incentives in Asia Table 1.9. Indexes of Comparative Advantage in Agriculture and Processed Food, Asian Focus Economies, 1965­2004 a. Revealed comparative advantage index (world 1.0)a Economy 1965­69 1975­79 1985­89 2000­04 East Asia 2.2 2.2 1.1 0.7 China 2.1 2.1 1.3 0.6 Indonesia 2.0 1.3 1.4 1.7 Korea, Rep. of 0.8 0.6 0.3 0.3 Malaysia 2.4 2.9 2.4 1.1 Philippines 3.1 2.8 1.9 0.7 Taiwan, China 1.5 0.6 0.4 0.3 Thailand 3.1 3.5 3.1 2.0 Vietnam -- -- -- 3.0 South Asia 1.9 2.0 1.8 1.4 Bangladesh -- 1.9 1.9 0.9 India 1.5 1.8 1.5 1.4 Pakistan 2.1 1.9 2.1 1.4 Sri Lanka 3.8 4.1 3.2 2.5 Total 2.2 2.2 1.2 0.8 b. Trade specialization index (world 0.0)b Economy 1965­69 1975­79 1985­89 2000­04 East Asia -- -- -- 0.03 China -- -- 0.07 0.16 Indonesia 0.48 0.42 0.43 0.16 Korea, Rep. of 0.63 0.41 0.45 0.53 Malaysia 0.44 0.60 0.56 0.29 Philippines 0.47 0.51 0.25 0.18 Taiwan, China 0.08 0.27 0.35 Thailand 0.68 0.69 0.57 0.44 Vietnam -- -- -- 0.44 South Asia -- 0.05 0.03 0.06 Bangladesh -- 0.37 0.46 0.62 India 0.18 0.13 0.16 0.10 Pakistan 0.20 0.13 0.05 0.24 Sri Lanka 0.34 0.30 0.21 0.08 Total -- -- -- 0.03 Sources: Sandri, Valenzuela, and Anderson 2007; World Development Indicators Database 2008. Note: -- no data are available. a. The share of agriculture and processed food in national exports as a ratio of such products in worldwide merchandise exports. The world average is 1.0. b. The ratio of net exports to the sum of the exports and imports of agricultural and processed food products. The world average is 0.0. Table 1.10. Export Orientation, Import Dependence, and Self-Sufficiency in Primary Agricultural Production, Asian Focus Economies, 1961­2004 (%, at undistorted prices) a. Exports, as a share of production Economy 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 0 0 0 0 0 0 0 2 1 Taiwan, China 5 9 13 14 10 10 6 5 6 China 2 2 2 3 5 5 7 7 7 Indonesia -- -- 6 5 5 6 4 5 4 Malaysia 70 64 54 41 35 34 19 12 9 Philippines 13 11 14 8 7 2 1 1 1 Thailand -- -- 13 20 24 26 25 25 30 Vietnam -- -- -- -- -- 3 4 9 11 Bangladesh -- -- -- 3 3 3 2 1 1 India 1 1 1 1 1 1 1 1 2 Pakistan 7 5 5 2 5 8 4 2 1 Sri Lanka 68 62 44 52 36 34 24 31 39 Total 3 4 4 4 4 6 6 5 5 (Table continues on the following page.) 19 20 Table 1.10. Export Orientation, Import Dependence, and Self-Sufficiency in Primary Agricultural Production, Asian Focus Economies, 1961­2004 (continued) b. Imports, as a share of apparent consumption Economy 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 4 5 12 8 11 9 11 11 13 Taiwan, China 24 33 56 66 76 81 86 90 93 China 2 2 2 3 5 5 7 7 7 Indonesia -- -- 0 1 1 1 1 2 2 Malaysia 13 6 3 1 1 1 2 3 6 Philippines 0 0 1 0 1 0 0 2 1 Thailand -- -- 0 0 0 0 0 2 5 Vietnam -- -- -- -- -- 0 0 0 0 Bangladesh -- -- -- 3 4 5 3 3 5 India 3 4 2 2 1 1 1 1 2 Pakistan 6 5 3 5 2 5 5 6 4 Sri Lanka 7 5 1 0 1 4 1 3 5 Total 3 4 4 5 7 12 17 17 19 c. Self-sufficiency ratio Economy 1961­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 96 95 88 92 89 91 90 91 87 Taiwan, China 80 73 51 40 27 21 15 11 7 China 99 101 100 99 98 101 101 99 98 Indonesia -- -- 106 105 104 106 104 103 102 Malaysia 293 265 215 167 152 150 122 110 104 Philippines 115 112 116 108 106 101 101 99 99 Thailand -- -- 115 125 131 135 133 130 137 Vietnam -- -- -- -- -- 103 104 110 112 Bangladesh -- -- 98 99 99 98 99 98 96 India 98 97 99 99 99 99 100 100 100 Pakistan 101 100 102 97 103 104 99 96 97 Sri Lanka 297 298 194 221 155 144 131 142 157 Total 100 100 100 99 97 94 88 87 85 Sources: Author compilation; FAO Agricultural Trade Database 2008. Note: Compiled using estimates of the total agricultural production of covered and noncovered products, valued at undistorted prices, plus total agricultural trade value data. Self-sufficiency is defined for each product as the ratio of production to production, plus imports, minus exports. -- no data are available. 21 22 Distortions to Agricultural Incentives in Asia trade measure is the sole government intervention, then the measured NRA will also be the consumer tax equivalent (CTE) rate at that same point in the value chain. Where there are also domestic producer or consumer taxes or subsidies, the NRA and CTE will no longer be equal, and at least one of them will be different from the price distortion at the border caused by trade measures.8 NRAs and CTEs may be used for several purposes, and the purpose affects the appropriate choice of methodology. In our project, we rely on NRAs and CTEs to achieve three purposes. One purpose is to generate a comparable set of numbers across a wide range of countries and over a long time period. This means the methodology must be both simple and somewhat flexible. Another purpose is to provide a single number, the NRA, to indicate the total extent of transfers to or from farmers because of government agricultural policies and another number, the CTE, to indicate the extent of the transfers to or from consumers. The NRA and the CTE are both expressed as a percentage of the undistorted price or in dol- lar terms. This is also the purpose of the OECD's producer and consumer support estimates, which may be negative if the transfers from the relevant group exceed the transfers to the relevant group. Our research project's agricultural NRAs and CTEs are similar in spirit to the OECD estimates, but there are also important dif- ferences, which are outlined below. Our third purpose is to enable economic mod- elers to use the NRAs as producer price wedges for individual primary and lightly processed agricultural products and to use the CTEs as consumer price wedges in single-sector, multisector, and economy-wide policy simulation models by allo- cating these wedges to particular policy instruments such as trade taxes or domes- tic producer or consumer subsidies or taxes. The NRAs are based on our estimates of assistance to individual industries. Great care has gone into generating the NRAs for each covered agricultural indus- try, particularly in countries where trade costs are high, the pass-through along the value chain is affected by imperfect competition, and the markets for foreign currency have been highly distorted to varying degrees at various times. Most distortions in industries producing tradables arise from trade measures such as quantitative trade restrictions or tariffs imposed on the import cost, insurance, and freight price or export subsidies or taxes imposed on the free on board price at the country's border. An ad valorem tariff or export subsidy is the equivalent of a production subsidy and a consumption tax at the same rate expressed as a percentage of the border price. For this reason, such tariffs and sub- sidies are captured in the NRAs and CTEs at the point in the value chain at which a product is first traded. To obtain the NRAs for farmers, the authors of the coun- try studies have estimated or guessed the extent of pass-through back to the farm- gate and added any domestic farm output subsidies. To obtain the CTEs, they have Introduction and Summary 23 also added any product-specific domestic consumer taxes to the distortions caused by border measures. Note that the NRAs and CTEs differ from the OECD's producer and consumer support estimates in that the latter pair is expressed as a percentage of the distorted price and, hence, will be lower (for positive protection rates) than the former pair, which is expressed as a percentage of the undistorted price. We have decided not to seek estimates of the more complex effective rate of assistance even though it is, in principle, a better single partial equilibrium measure of the distortions in producer incentives. To establish these alternative estimates, we must know, for each product, the value added and various intermediate input shares in output. These data are not available in most developing countries even for a few years, let alone for every year in the long time series that is the focus of our study. Moreover, in most countries, the distortions in farm input prices are small compared with the distortions in farm output prices. Nonetheless, if the product- specific distortions to input costs are significant, they may be captured by estimat- ing their equivalent values in terms of a higher output price and then including this estimate in the NRA for individual agricultural industries wherever data allow. We also add any non-product-specific distortions, including distortions in farm input prices, into the estimate for the overall sectoral NRA for agriculture. The targeted minimum product coverage of our NRA estimates is 70 percent based on the gross value of farm production at undistorted prices. This target cov- erage is similar to the coverage of the OECD producer support estimates. Unlike the OECD, however, we do not routinely assume that the nominal assistance for covered products applies equally to the farm products we do not cover. This is because, in developing countries, agricultural policies affecting our noncovered products are often different from those affecting our covered products. For exam- ple, nontradables among noncovered farm goods--often highly perishable prod- ucts or products of low value relative to the transport cost--are frequently not subject to direct distortionary policies. We have asked the authors of the country case studies to provide three sets of NRA guesstimates for noncovered farm prod- ucts, one each for the import-competing, exportable, and nontradable subsectors. Weighted averages for all agricultural products have then been generated using the gross values of production at unassisted prices as weights. For countries that also provide non-product-specific agricultural subsidies or taxes--assumed to be shared on a pro rata basis between tradables and nontradables--or assistance decoupled from production, the net assistance from these measures is then added to the product-specific assistance to obtain NRAs for total agriculture. We apply the same procedure to obtain NRAs also for tradable agriculture for use in gener- ating RRAs (which are defined below). 24 Distortions to Agricultural Incentives in Asia How best to present regional aggregate NRA and RRA estimates depends on the purpose for which the averages are required. We generate weighted average NRAs for covered products for each country because only then are we able to add the NRAs for noncovered products and the non-product-specific assistance to obtain the NRAs for all agriculture. In averaging across countries, we consider each polity as an observation of interest. So, a simple average is meaningful for the purpose of political economy analysis. However, if one wishes to acquire a sense of the distortion in the agriculture of an entire region, one needs a weighted average. The weighted average NRAs for covered primary agriculture may be generated by multiplying the share of each primary industry in the gross value of production (valued at farmgate equivalent undistorted prices) by the corresponding NRAs and adding across industries.9 The overall sectoral rate, which we denote as NRAag, may be obtained by also adding the actual or assumed information for noncovered farm commodities and, where it exists, the aggregate value of non- product-specific and decoupled assistance in agriculture. A weighted average can be similarly generated for the tradables part of agricul- ture, including those industries producing products such as milk and sugar that require only light processing before they are traded. This is accomplished by assuming that the product's share of non-product-specific assistance equals the product's weight in the total. Call this NRAagt. In addition to the mean, it is important also to provide a measure of the dis- persion or variability of the NRA estimates across the covered products. The cost of government policy distortions in incentives in terms of resource misallocation tends to be greater, the greater the degree of substitution in production (Lloyd 1974). In the case of an agricultural sector that involves a use of farmland that is sector specific, but transferable among farm activities, the greater the variation of NRAs across industries within the sector, then the higher will be the welfare cost of the market interventions. A simple indicator of dispersion is the standard devi- ation in the NRAs of the covered industries. Each industry is classified as import-competing, or as a producer of exporta- bles, or as a producer of nontradables. This status sometimes changes over time. It is possible to generate for each year the weighted average NRAs for the two groups of tradables (import-competing products and exportables). These NRAs are used to generate a trade bias index (TBI), which is defined as follows: TBI [(1 NRAagx 100) (1 NRAagm 100) 1], (1.1) where NRAagm and NRAagx are the average percentage NRAs for the import- competing and exportable parts of the agricultural sector. The TBI indicates, in a single number, the extent to which the (typically) antitrade bias (negative TBI) in agricultural policies changes over time. Introduction and Summary 25 Farmers are affected not only by the prices for their own outputs, but also, albeit indirectly, through changes in factor market prices and the exchange rate because of the incentives nonagricultural producers face. Thus, it is the relative prices and, hence, the relative rates of government assistance that affect producer incentives. More than 70 years ago, Lerner (1936) provided the symmetry theo- rem, which proved that, in a two-sector economy, an import tax has the same effect as an export tax. This carries over to a model that also includes a third sec- tor producing only nontradables and to a model with imperfect competition, regardless of the economy's size (Vousden 1990). If one assumes that there are no distortions in the markets for nontradables and that the value shares of agricul- tural and nonagricultural nontradable products remain constant, then the economy-wide effect of distortions in agricultural incentives may be captured by measuring the extent to which the tradable parts of agricultural production are assisted or taxed relative to the producers of other tradables. Because we are able to generate estimates of the average NRAs for nonagricultural tradables, we are then able to calculate an RRA, which is defined in percentage terms as follows: RRA 100[(1 NRAagt 100) (1 NRAnonagt 100) 1], (1.2) where NRAagt and NRAnonagt are the weighted average percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. Since the NRA cannot be less than 100 percent if producers are to earn anything, nei- ther can the RRA (assuming NRAnonagt is positive). If both of those sectors are equally assisted, the RRA is zero. This measure is useful in that, if it is below (above) zero, it provides an internationally comparable indication of the extent to which a country's policy regime has an anti- or proagricultural bias. In calculating the NRAs for producers of agricultural and nonagricultural tradables, we use the methodology outlined in appendix A to seek to include distortions generated by dual or multiple exchange rates. Direct interventions in the markets for foreign currency were common in some Asian countries, includ- ing China in the 1970s and 1980s. However, authors of some of the focus coun- try studies have experienced difficulty in determining an appropriate estimate of the extent of this distortion. So, the impact of such interventions on the NRAs has not been included in all the studies. The exclusion of this impact in some countries (for example, India) means that the estimated (typically) posi- tive NRAs for importables and (typically) negative NRAs for exportables are smaller than they should be in these studies. In cases where the NRAs for importables dominate the NRAs for exportables, this omission would lead to an underestimate of the average (positive) NRAs for such tradables. The converse would also be true. In either case, this leads also to an underestimate of the (anti-)TBI. 26 Distortions to Agricultural Incentives in Asia To obtain U.S. dollar values for farmer assistance and consumer taxation, Valenzuela, Kurzweil, Croser, Nelgen, and Anderson (see appendix B) have multi- plied the country author NRA estimates by the gross value of production at undistorted prices to obtain an estimate, in current U.S. dollars, of the direct gross subsidy equivalents of assistance to farmers. These estimates are then added across the products of a country and across countries for all products to obtain regional aggregate transfer estimates for the economies under study. These values are cal- culated in constant dollars and are also expressed as estimates per farmworker. To obtain comparable U.S. dollar value estimates of consumer transfers, the CTE estimate at the point at which a product is first traded is multiplied by con- sumption (obtained from the FAO SUA-FBS Database), which is valued at undis- torted prices, to obtain an estimate in constant U.S. dollars of the tax equivalent to consumers of primary farm products. This is then added across the products of a country and across countries for any or all products to obtain regional aggregate transfer estimates on the economies under study. These values are also expressed on a per capita basis. Estimates of Asian Policy Indicators We begin with the NRAs in agriculture and compare them with the NRAs for nonagricultural tradables. We also present constant U.S. dollar equivalents of the assistance and taxation among farmers, as well as the CTEs of policies as they affect the buyers of farm products. NRAs in agriculture Agricultural price and trade policies reduced average farmer earnings in develop- ing Asia throughout the period up to the 1980s by more than 20 percent.10 This implicit taxation declined beginning in the early 1980s, and, in the mid-1990s, the NRAs switched sign and became increasingly positive. The averages hide consid- erable diversity within the region, however. The nominal assistance for farmers in Korea and Taiwan, China was positive starting in the early 1960s (although small initially). The NRAs were slightly above zero in Indonesia during some years in the 1970s and 1980s, as they were in Pakistan prior to the independence of Bangladesh (East Pakistan) in 1971. The average NRAs in India and the Philip- pines became positive in the 1980s (table 1.11). A visual impression of the differ- ences across countries and the rise in average NRAs is supplied in figure 1.2, in which the situation in 2000­04 is compared with the situation in 1980­84. This trend is evident among the vast majority of the commodity NRAs for the region, too. Meat and milk were the only products for which the assistance rates were cut over 1980­2004 (figure 1.3). Table 1.11. NRAs for Agricultural Products, Asian Focus Economies, 1955­2004 (percent) Economy, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Northeast Asia 42.8 42.6 41.7 41.2 39.5 38.2 25.7 1.7 14.4 11.9 Korea, Rep. of 3.2 4.0 13.4 35.7 56.3 89.4 126.1 152.8 129.8 137.3 Taiwan, Chinaa 12.0 3.6 3.0 9.3 7.1 14.9 27.1 38.1 46.4 61.3 Chinab 45.2 45.2 45.2 45.2 45.2 45.2 35.5 14.3 6.6 5.9 Southeast Asia -- 6.8 5.9 8.8 0.0 4.6 0.4 4.2 0.0 11.1 Indonesia -- -- -- 2.6 9.3 9.2 1.7 6.6 8.6 12.0 Malaysia -- 7.2 7.5 9.0 13.0 4.6 1.3 2.3 0.2 1.2 Philippines -- 5.3 14.4 5.1 7.1 1.0 18.7 18.5 32.9 22.0 Thailand -- -- -- 20.3 14.0 2.0 6.2 5.7 1.7 0.2 Vietnama -- --- -- -- -- -- 13.9 25.4 0.6 21.2 South Asia 0.0 0.5 0.6 0.4 5.5 0.6 20.9 0.7 0.2 13.6 Bangladesh -- -- -- 16.0 1.4 3.3 11.7 1.5 5.2 2.7 Indiab 0.1 0.1 0.1 0.2 5.6 1.9 24.9 1.8 0.7 15.8 Pakistana -- 0.7 15.3 6.8 8.5 6.4 4.0 6.9 1.6 1.2 Sri Lanka 2.3 22.8 24.5 16.3 25.5 13.5 9.9 1.2 12.2 9.5 Unweighted averagec 12.3 10.2 3.9 4.7 4.1 3.2 11.5 12.1 16.8 21.7 Weighted average 27.3 26.7 25.1 25.3 23.8 20.6 9.0 2.0 7.5 12.0 Dispersion, country NRAsd 25.3 25.2 31.2 29.9 32.5 39.9 43.8 47.5 36.6 38.0 Product coveragee 52 65 63 64 68 74 78 75 73 66 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: For each economy, the figure shows weighted averages, including product-specific input distortions, non-product-specific assistance, and guesstimates by the authors of chapters 2­12 of the averages for noncovered farm products. The weights are based on the gross value of total agricultural production at undistorted prices. -- no data are available. a. Pakistan: 1960-64 is 1961-64. Taiwan, China: 2000-04 is 2000-03. Vietnam: 1985-89 is 1986-89. b. The estimates for China before 1981 and India before 1965 are based on the assumption that the NRAs were the same as the average NRAs for those economies in 1981­84 and 1965-69, respectively, and that the gross value of production of those economies in the missing years is the same as the average share of the value of the production of the economies in total world production in 1981-84 and 1965-69, respectively. This assumption is conservative in that, in both countries, the average NRAs were probably even lower in earlier years. 27 c. The unweighted average is the simple average across the 12 economies of the national weighted NRA averages. d. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the national NRAs. e. Weighted averages for the covered products. 28 Distortions to Agricultural Incentives in Asia Figure 1.2. NRAs in Agriculture, Asian Focus Economies, 1980­84 and 2000­04 160 140 120 100 80 % 60 40 NRA, 20 0 20 40 60 of Rep. China India Lanka China Vietnam Sri MalaysiaPakistanThailand Philippines Indonesia Bangladesh Korea, Taiwan, economy 1980­84 2000­04 Sources: Author compilation; Anderson and Valenzuela, forthcoming; estimates reported in chapters 2­12 of this volume. Note: There are no data for Vietnam in 1980­84. The estimates for China before 1981 are based on the assumption that the NRAs were the same as the average NRAs in 1981­84 and that the gross value of production in the missing years is the same as the average share of the value of production of the economy in total world production in 1981­84. Figure 1.3 also illustrates the diversity of the region's average NRAs across farm commodities. In Asia, as in other regions, the product NRAs for rice pudding-- rice, milk, and sugar--are among the highest, but, even among these three prod- ucts, there is great diversity across economies in the NRAs. The five-year averages range from almost zero to as much as 400 percent for rice and 140 percent for milk in Korea and to 230 percent for sugar in Bangladesh (figure 1.4). This sug- gests that the production of these products in Asia is not optimally allocated for efficient resource use. There is a great deal of diversity across economies in the average NRAs and across commodities within the farm sector of each economy. Measured through the standard deviation, the extent of both types of diversity has grown rather than diminished over the past five decades. The cross-economy diversity of average Figure 1.3. NRAs, by Product, Asian Focus Economies, 1980­84 and 2000­04 a. Unweighted average soybeans sugar poultry rice milk pig meat wheat product beef maize cotton rubber coconuts palm oil 50 0 50 100 150 NRA, % 2000­04 1980­84 b. Weighted averagea beef sugar milk rice soybeans maize poultry product wheat cotton pig meat rubber palm oil coconuts 50 0 50 100 150 NRA, % 2000­04 1980­84 Sources: Author compilation; Anderson and Valenzuela, forthcoming; estimates reported in chapters 2­12 of this volume. a. Weights are based on the gross value of total agricultural production at undistorted prices. Each NRA by economy and by product has been weighted by the value of the production in the economy of that commodity in a given year. Products representing less than 1 percent of the gross value of regional production are excluded. These include cocoa, onions, chilies, barley, jute, sunflower seeds, garlic, peppers, cabbages, cassava, potatoes, eggs, tea, coffee, sorghum, rapeseeds, chickpeas, groundnuts, and beef. 30 Distortions to Agricultural Incentives in Asia Figure 1.4. NRAs for Rice, Milk, and Sugar, Asian Focus Economies, 1980­84 and 2000­04 a. Rice 400 300 200 % NRA, 100 0 100 of India Rep. China Lanka China MalaysiaPhilippines Vietnam IndonesiaSri ThailandPakistan Bangladesh Korea, Taiwan, economy 1980­84 2000­04 b. Milk 300 200 % NRA, 100 0 Korea, Rep. of India China Pakistan economy 1980­84 2000­04 Introduction and Summary 31 Figure 1.4. (continued) c. Sugar 300 200 % NRA, 100 0 India China Vietnam Pakistan Thailand Bangladesh Philippines Indonesia economy 1980­84 2000­04 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: Vietnam: 1980­84 is 1986­89 for rice and 1990­95 for sugar. NRAs is evident at the bottom of table 1.11: the dispersion indicator rises from around 25 percent in the early years we are studying to around 40 percent in recent years. The cross-commodity diversity within each economy is clear in table 1.12 (unweighted averages), where the standard deviation among NRAs is reported for covered products (which account for up to 75 percent of the value of agricultural production at undistorted prices). This means that much could be gained through improved resource reallocation among economies and within the agricultural sector of each economy if the differences in rates of assistance were reduced. A striking feature of the distortion pattern within the farm sector is the strong antitrade bias. This is evident in figure 1.5, which depicts the average NRAs in the region for agriculture's import-competing and export subsectors. The average NRAs for import-competing products are always positive, and the trend is upward-sloping, whereas the average NRAs for exportables are negative before gradually approaching zero after the 1980s. While the gap between the NRAs for these two subsectors has diminished little in the region since the 1960s, there are nonetheless several countries--Malaysia, Pakistan, Sri Lanka, Thailand--in which the gap has narrowed and, hence, the TBIs have approached zero (table 1.13). The rise in the average NRAs since the 1980s is too large to be explained only by loses in the comparative advantage of farm products as economies industrial- ize. The export shares indicated in table 1.13 show that the share of the export (Text continues on page 37.) 32 Table 1.12. NRA Dispersion across Covered Agricultural Products, Asian Focus Economies, 1955­2004 (percent) Economy 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Northeast Asiaa 38.5 38.5 66.8 61.7 68.6 64.0 67.1 80.5 102.3 116.6 Korea, Rep. of 34.1 40.5 85.0 82.5 89.0 80.1 114.8 164.2 200.1 225.4 Taiwan, Chinab 42.8 36.4 48.7 40.9 48.2 37.5 34.3 56.5 88.4 109.0 China -- -- -- -- -- 74.3 52.3 20.7 18.4 15.3 Southeast Asiaa -- 21.1 30.4 23.9 32.7 46.2 47.9 42.5 39.7 40.4 Indonesia -- -- -- 29.1 49.4 53.6 35.0 40.5 49.0 33.3 Malaysia -- 19.8 10.6 11.0 34.8 58.6 90.5 71.9 33.7 40.1 Philippines -- 22.5 50.2 28.3 24.2 42.7 50.9 30.1 40.6 37.6 Thailand -- -- -- 27.5 22.4 30.1 29.3 25.1 22.9 16.7 Vietnamb -- -- -- -- -- -- 33.6 44.7 52.4 74.3 South Asiaa 44.2 46.3 64.9 51.8 55.2 51.5 86.2 55.4 45.0 58.7 Bangladesh -- -- -- -- 71.4 67.6 190.7 77.5 67.9 101.2 India -- 35.5 68.0 45.8 49.8 39.2 46.9 28.5 19.4 21.5 Pakistanb -- 74.6 105.8 77.6 45.0 52.2 69.4 34.4 29.4 43.1 Sri Lanka 44.2 28.7 20.9 31.9 54.6 46.9 37.9 81.3 63.3 69.1 Unweighted averagea 39.0 36.8 55.5 41.5 48.0 51.1 66.9 55.9 55.6 64.4 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: Dispersion for each country is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs across covered products. -- no data are available. a. The unweighted average is the simple average across the relevant economies of the simple five-year dispersion measure averages. b. Pakistan: 1960­64 is 1961­64. Taiwan, China: 2000­04 is 2000­03. Vietnam: 1985­89 is 1986­89. Figure 1.5. NRAs for Exportable, Import-Competing, and All Agri- cultural Products, Asian Focus Economies, 1955­2004 a. Unweighted averages 40 30 20 10 % 0 NRA, 10 20 30 40 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year import-competing products exportables total b. Weighted averages 50 30 10 % 0 NRA, 10 30 50 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year import-competing products exportables total Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: Total NRAs may be above or below the exportable and importable averages because assistance to nontradables and non-product-specific assistance are also included. The values used in the NRA estimates are based on the assumption that the NRAs in agriculture in China before 1981 and in India before 1965 are the same as the average NRAs in those economies in 1981­84 and 1965­69, respectively, and that the gross value of production of those economies in the missing years is the same as the average share of the value of the production of the economies in total world production in 1981­84 and 1965­69, respectively. Table 1.13. NRAs for Agricultural Exportable and Import-Competing Products and the TBI, Asian Focus 34 Economies, 1955­2004 (percent) Economy, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of NRA, exportable -- -- -- -- -- -- -- -- -- -- NRA, import-competing 3.3 4.9 16.3 46.1 71.8 118.6 159.8 197.6 164.8 171.9 TBIa 0.03 0.05 0.14 0.32 0.42 0.54 0.62 0.66 0.62 0.63 Export shareb 0 0 0 0 0 0 0 0 0 0 Taiwan, Chinac NRA, exportable 18.1 5.7 4.3 15.4 10.3 25.1 48.9 57.1 57.0 70.3 NRA, import-competing 3.6 0.5 2.3 3.3 5.2 9.2 18.1 42.0 54.3 71.3 TBIa 0.15 0.05 0.02 0.12 0.05 0.15 0.27 0.11 0.02 0.00 Export shareb 85 80 78 74 70 61 47 28 25 24 China NRA, exportable -- -- -- -- -- 56.9 46.0 21.8 0.8 0.1 NRA, import-competing -- -- -- -- -- 11.0 20.4 2.2 17.0 7.3 TBIa -- -- -- -- -- 0.50 0.55 0.23 0.15 0.07 Export shareb -- -- -- -- -- 79 88 80 75 72 Northeast Asia NRA, exportable 18.1 5.7 4.3 15.4 10.3 40.0 44.2 20.5 0.2 0.2 NRA, import-competing 3.4 3.9 14.4 38.3 59.7 21.2 51.8 42.8 40.5 26.4 TBIa 0.15 0.02 0.09 0.17 0.31 0.50 0.63 0.44 0.29 0.21 Export shareb 35 31 32 29 29 76 83 74 70 69 Indonesia NRA, exportable -- -- -- 3.3 0.3 7.0 16.5 24.6 17.2 3.3 NRA, import-competing -- -- -- 1.3 16.5 19.5 5.1 0.7 5.8 24.7 TBIa -- -- -- 0.01 0.14 0.21 0.20 0.24 0.12 0.22 Export shareb -- -- -- 57 37 34 32 29 34 37 Malaysia NRA, exportable -- 11.4 9.0 12.7 18.7 11.8 5.0 4.1 3.0 1.3 NRA, import-competing -- 13.0 1.8 2.5 21.6 36.7 44.1 33.2 10.8 12.3 TBIa -- 0.22 0.06 0.14 0.31 0.35 0.33 0.28 0.12 0.12 Export shareb -- 84 79 78 86 86 86 82 80 81 Philippines NRA, exportable -- 6.2 35.4 10.2 9.9 3.8 6.2 4.8 0.7 3.7 NRA, import-competing -- 5.3 10.1 2.7 6.1 0.6 30.0 27.6 47.9 30.8 TBIa -- 0.00 0.23 0.04 0.04 0.04 0.18 0.17 0.32 0.25 Export shareb -- 21 22 30 41 41 39 33 25 22 Thailand NRA, exportable -- -- -- 26.7 19.4 11.1 11.7 9.2 3.8 0.6 NRA, import-competing -- -- -- 4.8 1.9 45.3 22.0 6.4 34.4 4.7 TBIa -- -- -- 0.18 0.20 0.37 0.24 0.14 0.27 0.03 Export shareb -- -- -- 83 83 84 84 81 85 86 Vietnamc NRA, exportable -- -- -- -- -- -- 17.5 27.1 2.0 17.7 NRA, import-competing -- -- -- -- -- -- 37.1 25.8 65.4 67.3 TBIa -- -- -- -- -- -- 0.37 0.42 0.40 0.29 Export shareb -- -- -- -- -- -- 98 96 96 94 Southeast Asia NRA, exportable -- 10.8 4.4 13.9 10.8 8.7 11.0 14.9 7.6 1.0 NRA, import-competing -- 4.5 8.2 2.7 11.1 18.4 10.6 5.8 8.0 24.9 TBIa -- 0.15 0.03 0.12 0.20 0.23 0.19 0.20 0.14 0.19 Export shareb -- 57 44 61 51 48 51 48 51 53 Bangladesh NRA, exportable -- -- -- -- 34.6 26.2 32.4 33.0 9.9 33.2 NRA, import--competing -- -- -- -- 6.5 1.9 24.4 0.1 7.9 6.0 TBIa -- -- -- -- 0.30 0.23 0.45 0.33 0.00 0.37 Export shareb -- -- -- -- 13 9 7 11 7 5 India NRA, exportable -- -- 37.4 22.3 35.9 27.8 6.0 15.3 12.4 6.4 NRA, import-competing -- -- 41.7 52.7 74.5 58.8 81.4 38.3 22.5 34.2 35 TBIa -- -- 0.55 0.50 0.63 0.54 0.47 0.38 0.28 0.30 Export shareb -- -- 58 74 73 63 43 70 71 51 (Table continues on the following page.) Table 1.13. NRAs for Agricultural Exportable and Import-Competing Products and the TBI, Asian Focus 36 Economies, 1955­2004 (continued) Economy, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Pakistanc NRA, exportable -- 33.3 35.3 20.1 33.5 29.1 32.1 16.7 4.4 5.6 NRA, import-competing -- 1.8 45.0 19.2 4.3 1.9 5.4 7.9 1.9 3.7 TBIa -- 0.38 0.55 0.27 0.31 0.28 0.35 0.10 0.02 0.08 Export shareb -- 31 29 24 25 27 30 24 21 22 Sri Lanka NRA, exportable 22.8 40.0 38.6 41.1 45.2 31.1 21.4 24.2 2.0 5.9 NRA, import-competing 62.5 11.9 5.9 9.0 3.7 0.6 2.1 22.4 31.8 12.8 TBIa 0.52 0.45 0.35 0.45 0.43 0.31 0.18 0.38 0.25 0.05 Export shareb 76 72 68 50 62 58 52 51 58 58 South Asia NRA, exportable 22.8 37.5 37.2 30.0 36.1 27.9 20.6 15.8 12.0 6.2 NRA, import-competing 62.5 39.2 41.2 39.4 45.1 37.9 63.3 25.1 14.5 26.5 TBIa 0.52 0.55 0.56 0.50 0.56 0.48 0.51 0.33 0.23 0.26 Export shareb 76 27 53 61 63 56 32 63 64 46 Asia, unweighted averaged NRA, exportable 20.4 16.3 12.5 15.5 20.8 17.3 12.5 10.4 0.1 0.9 NRA, import-competing 18.9 6.8 15.2 13.2 17.7 24.6 36.4 31.4 34.5 34.5 TBI 0.33 0.22 0.24 0.25 0.33 0.34 0.36 0.32 0.26 0.25 Asia, weighted averaged NRA, exportable 20.2 12.6 27.2 20.7 25.4 41.7 38.1 19.1 4.3 0.6 NRA, import-competing 5.9 4.7 33.9 26.6 31.3 20.8 43.8 25.8 24.8 25.4 TBI 0.25 0.17 0.46 0.37 0.43 0.52 0.57 0.36 0.23 0.21 Export shareb 47 37 50 58 55 65 69 66 65 63 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: -- no data are available. a. The TBI is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. b. The export share is the share of exportables in the gross value of the production of all agricultural tradables at undistorted prices. c. Pakistan: 1960­64 is 1961­64. Taiwan, China: 2000­04 is 2000­03. Vietnam: 1985­89 is 1986­89. d. These NRAs differ from those in figure 1.5 because these do not include backcast estimates for China and India. The regional TBI averages are calculated from the regional NRA averages for the exportable and import-competing parts of the agricultural sector. Introduction and Summary 37 subsector in total tradable farm production has not declined much in most Asian economies. This suggests that the main motive for intervention cannot have been to raise revenue by taxing trade; nor can it have only been to reduce distortions. Otherwise, there would have been no overshooting in the transition from taxing to assisting farmers, on average, nor would there have been the increase in assis- tance to import-competing farmers. The contributions to overall agricultural NRAs by input subsidies, domestic output taxes or subsidies, and trade measures at the border are summarized in table 1.14. (Non-product-specific assistance, which is not shown in the table, added only around 2 percentage points during the period of our study.) The dis- tortions in product-specific input prices contributed little, on average, to the over- all regional NRAs in agriculture. They reduced the negative value slightly in the 1960s and have added slightly to the positive value during the past decade or so. The contribution of input subsidies is largest in India, especially in recent years when it has added almost 10 percentage points to the sector's NRAs and has been a nontrivial item in overall government budgets. Earlier, the contribution of input subsidies was high in Indonesia, too, peaking at 7 percentage points in the 1980s, before reforms sent them back to only 2 percentage points more recently. There are fewer domestic output subsidies, on average, in the region now, although such subsidies were implemented in India and Sri Lanka during earlier decades (table 1.14). In China, output taxes in kind were prevalent until recently. The U.S. dollar value equivalents of the positive or negative assistance to farm- ers because of agricultural price and trade policies have been nontrivial. The anti- agricultural bias peaked in the region in the late 1970s at more than US$130 bil- lion per year at constant 2000 U.S. dollars (2000 U.S. dollars; see the bottom row of table 1.15, panel a). This is equivalent to a gross tax of around US$170 for each person engaged in agriculture. Most of the US$130 billion was generated by China's antiagricultural policies. India was the second most important contribu- tor in the 1960s and 1970s. Thanks to the reforms of the past two decades, this taxation has gradually disappeared in all our focus economies. This reform does not mean there is no intervention now. Rather, the negative influence has been replaced by positive assistance to farmers in most countries, totaling around US$60 billion per year in recent years. China, India, and Korea each contributed about one-quarter of the total. In recent years, the total has averaged around US$60 per farmworker in Asia (table 1.15, panel b). This is not insignificant rela- tive to per capita income in the region, but the amount is unevenly distributed. It ranges from around US$6,900 and US$5,300 in Korea and Taiwan, China (more than one-third of per capita income in those economies) to around $150 in the Philippines, $90 in Indonesia, $60 in India, Malaysia, and Vietnam, and virtually zero in Bangladesh, Pakistan, and Thailand. (Text continues on page 44.) 38 Table 1.14. NRAs for Covered Agricultural Products, by Policy Instrument, Asian Focus Economies, 1955­2004 (percent) Economy, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of NRA, inputs 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, domestic MSa 0.2 0.4 0.9 4.2 7.1 5.3 5.5 5.9 6.1 5.2 NRA, border MSa 3.7 4.7 15.7 43.4 66.7 117.5 161.2 196.0 176.9 208.5 NRA, total agriculture 3.9 4.4 16.6 47.6 73.8 122.8 166.7 201.9 182.9 213.6 Taiwan, Chinab NRA, inputs 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, domestic MSa 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, border MSa 23.6 7.3 6.3 20.4 14.4 36.9 82.4 117.4 142.2 178.4 NRA, total agriculture 23.6 7.3 6.3 20.4 14.4 36.9 82.4 117.4 142.2 178.4 China NRA, inputs -- -- -- -- -- 0.3 0.3 0.2 0.7 0.5 NRA, domestic MSa -- -- -- -- -- 12.6 6.3 6.2 1.1 1.4 NRA, border MSa -- -- -- -- -- 38.5 34.6 12.9 2.7 1.8 NRA, total agriculture -- -- -- -- -- 50.8 40.6 18.9 2.3 0.9 Northeast Asiac NRA, inputs 0.0 0.0 0.0 0.0 0.0 0.2 0.3 0.2 0.7 0.5 NRA, domestic MSa 0.1 0.2 0.6 2.6 4.6 8.4 5.8 5.5 0.8 1.2 NRA, border MSa 12.1 5.7 12.3 34.9 48.3 11.4 26.6 1.5 10.9 9.0 NRA, total agriculture 12.3 5.5 12.9 37.5 53.0 19.6 32.2 6.8 10.7 8.3 Indonesia NRA, inputs -- -- -- 6.0 6.8 7.3 6.0 3.2 2.6 2.3 NRA, domestic MSa -- -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, border MSa -- -- -- 8.7 4.3 4.8 6.3 8.8 11.7 13.3 NRA, total agriculture -- -- -- 2.7 11.1 12.2 0.3 5.5 9.1 15.6 Malaysia NRA, inputs -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, domestic MSa -- 9.0 8.1 10.2 17.0 6.9 0.4 2.5 0.1 1.5 NRA, border MSa -- 0.6 0.6 0.3 1.8 1.3 1.3 1.0 0.3 0.6 NRA, total agriculture -- 8.4 8.7 10.5 15.3 5.7 1.8 3.4 0.3 2.1 Philippines NRA, inputs -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, domestic MSa -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, border MSa -- 6.7 16.7 6.0 8.7 1.6 21.9 21.4 37.8 24.9 NRA, total agriculture -- 6.7 16.7 6.0 8.7 1.6 21.9 21.4 37.8 24.9 Thailand NRA, inputs -- -- -- 1.3 1.2 1.1 2.7 1.8 1.4 0.8 NRA, domestic MSa -- -- -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, border MSa -- -- -- 24.5 17.2 7.3 7.0 5.9 0.3 0.2 NRA, total agriculture -- -- -- 25.8 18.4 8.4 9.7 7.7 1.1 0.6 Vietnamb NRA, inputs -- -- -- -- -- -- 0.0 0.0 0.0 0.0 NRA, domestic MSa -- -- -- -- -- -- 0.7 0.9 0.0 1.7 NRA, border MSa -- -- -- -- -- -- 12.6 24.6 0.6 19.0 NRA, total agriculture -- -- -- -- -- -- 13.3 25.4 0.7 20.6 Southeast Asiac NRA, inputs -- 0.0 0.0 2.1 3.5 4.0 2.7 1.3 1.0 0.9 NRA, domestic MSa -- 6.3 3.3 1.4 2.1 0.9 0.0 0.1 0.0 0.4 NRA, border MSa -- 1.8 9.6 11.2 2.1 1.5 2.2 4.5 0.1 11.9 NRA, total agriculture -- 8.1 6.4 10.5 0.6 4.7 0.5 3.0 0.9 13.1 Bangladesh NRA, inputs -- -- -- -- 1.1 1.2 1.2 1.5 2.2 2.6 NRA, domestic MSa -- -- -- -- 0.0 0.0 0.0 0.0 0.0 0.0 NRA, border MSa -- -- -- -- 1.7 4.9 15.5 3.7 9.8 1.3 NRA, total agriculture -- -- -- -- 2.8 3.8 16.8 2.2 7.6 3.9 India NRA, inputs -- -- 0.0 0.0 0.0 0.7 4.4 5.7 7.2 9.5 NRA, domestic MSa -- -- 18.1 17.8 3.7 2.1 4.3 3.4 0.1 0.2 39 NRA, border MSa -- -- 17.9 17.7 9.3 0.8 16.2 7.3 6.4 6.1 NRA, total agriculture -- -- 0.3 0.2 5.6 1.9 24.9 1.8 0.7 15.8 (Table continues on the following page.) Table 1.14. NRAs for Covered Agricultural Products, by Policy Instrument, Asian Focus Economies, 1955­2004 (continued) 40 Economy, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Pakistanb NRA, inputs -- 4.5 2.1 1.6 4.6 3.5 2.9 2.3 1.9 1.4 NRA, domestic MSa -- 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, border MSa -- 1.3 23.9 10.9 16.4 12.8 8.8 12.5 4.5 0.1 NRA, total agriculture -- 5.8 21.7 9.3 11.8 9.3 5.9 10.2 2.6 1.5 Sri Lanka NRA, inputs 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 NRA, domestic MSa 4.2 8.1 6.9 5.0 5.1 5.0 3.6 1.5 0.7 0.9 NRA, border MSa 14.5 38.0 36.9 25.4 37.0 24.2 16.3 3.1 10.8 7.8 NRA, total agriculture 10.3 29.9 30.0 20.3 31.9 19.2 12.6 1.7 11.5 8.6 South Asiac NRA, inputs 0.0 1.5 0.2 0.1 0.4 0.9 4.0 5.0 6.3 8.3 NRA, domestic MSa 4.2 8.3 16.3 16.1 3.1 1.8 3.6 2.8 0.1 0.2 NRA, border MSa 14.5 14.7 14.9 15.4 9.2 2.5 13.5 7.5 6.2 5.2 NRA, total agriculture 10.3 7.8 1.2 0.6 5.7 0.2 21.0 0.3 0.0 13.6 Asia, unweighted average NRA, inputs 0.0 0.6 0.3 0.3 1.1 1.1 1.0 0.9 1.1 1.3 NRA, domestic MSa 1.4 0.6 2.4 1.4 0.8 0.9 0.1 0.0 0.1 0.2 NRA, border MSa 13.4 6.1 1.1 0.9 0.3 6.5 17.7 20.8 26.2 30.9 NRA, total agriculture 12.0 6.2 3.2 0.9 0.6 6.6 18.8 21.7 27.3 32.4 Asia, weighted averaged NRA, inputs 0.0 0.5 0.1 0.4 1.4 1.1 1.5 1.7 2.3 2.6 NRA, domestic MSa 1.4 2.0 13.0 10.6 1.0 5.3 2.9 2.5 0.6 0.7 NRA, border MSa 12.4 4.2 10.2 10.8 2.8 17.4 13.7 4.0 4.3 8.2 NRA, total agriculture 11.1 2.7 2.6 0.3 0.4 21.6 15.2 4.8 6.0 10.2 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: -- no data are available. a. MS market support, which is provided through domestic subsidies (or taxes) or through a border measure such as an import tariff (or subsidy) or an export subsidy (or tax). b. Pakistan: 1960­64 is 1961­64. Taiwan, China: 2000­04 is 2000­03. Vietnam: 1985­89 is 1986­1989. c. The weights for the subregional averages are based on the gross value of agricultural production at undistorted prices. d. The weight is the gross value of the production of covered products at undistorted prices. Table 1.15. Gross Subsidy Equivalents of Agricultural Assistance, Total and Per Farm Worker, Asian Focus Economies, 1955­2004 a. Total (at constant 2000 US$, millions, using the U.S. GDP deflator) Economy 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 154 107 854 1,672 6,943 9,335 13,306 18,594 17,536 15,289 Taiwan, Chinaa 394 133 132 439 605 1,342 2,500 3,849 4,170 3,725 Chinab 52,857 69,648 70,671 98,931 124,086 118,224 75,780 28,381 15,667 15,644 Indonesia -- -- -- 848 3,783 4,131 785 2,729 4,101 4,286 Malaysia -- 250 246 547 1,097 456 75 156 3 100 Philippines -- 225 735 1,082 903 299 1,399 1,850 3,832 1,951 Thailand -- -- -- 2,434 2,148 324 645 719 260 14 Vietnama -- -- -- -- -- -- 726 1,815 18 1,602 Bangladesh -- -- -- -- 583 672 882 103 448 189 Indiab 46 61 993 7,803 8,653 49 21,607 1,600 281 15,433 Pakistana -- 91 1,089 34 815 787 380 755 260 95 Sri Lanka 68 461 455 396 571 344 194 27 245 154 Total 53,426 70,373 69,554 109,965 126,359 106,348 38,740 8,481 37,169 58,455 All Asian economiesc 56,836 74,865 73,994 116,984 134,424 113,136 41,213 9,023 39,541 62,186 (Table continues on the following page.) 41 42 Table 1.15. Gross Subsidy Equivalents of Agricultural Assistance, Total and Per Farm Worker, Asian Economies, 1955­2004 (continued) b. Per person engaged in agriculture (at constant 2000 US$ using the U.S. GDP deflator) Economy 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 20 155 293 1,196 1,716 3,041 5,618 6,445 6,899 Taiwan, Chinaa 76 76 261 390 1,045 2,077 3,699 4,795 5,329 Chinab 235 222 281 319 280 163 57 31 31 Indonesia -- -- 27 113 113 19 60 86 86 Malaysia 135 126 267 515 213 36 79 2 56 Philippines 33 99 132 99 30 132 163 318 155 Thailand -- -- 163 130 18 34 36 13 1 Vietnama -- -- -- -- -- 33 73 1 57 Bangladesh -- -- -- 20 22 26 3 12 5 Indiab 0 6 43 43 0 97 7 1 57 Pakistana 7 78 2 47 41 19 35 11 4 Sri Lanka 217 195 155 207 116 60 8 66 40 Total 125 115 166 174 136 46 9 40 61 All Asian economiesc 125 115 166 174 136 46 9 40 61 c. By subsector (at undistorted farmgate prices, constant US$, billions) All Asian economies, from direct Focus economies assistance to farmerse Guesstimate, Covered Noncovered Non-product- other Asian Import-competing Year productsd farm productsd specific assistance economiesc Total Exportables products 1955­59 0.7 0.1 0.4 56.6 56.8 58.6 10.1 1960­64 0.8 0.1 1.0 74.4 74.7 74.4 12.0 1965­69 1.4 0.3 0.8 74.4 74.0 78.6 14.8 1970­74 7.7 3.7 0.3 106.3 117.4 113.9 11.1 1975­79 0.9 1.4 0.7 131.4 134.4 142.9 26.5 1980­84 76.4 6.2 14.0 44.6 113.1 140.3 26.6 1985­89 50.0 6.2 15.4 12.8 41.1 97.1 48.4 1990­94 14.9 1.1 4.4 0.3 9.1 49.5 33.7 1995­99 21.7 8.1 0.1 9.8 39.5 13.4 41.1 2000­04 32.0 14.3 1.5 15.8 60.6 2.0 42.0 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: -- no data are available. a. Pakistan: 1960­64 is 1961­64. Taiwan, China: 2000­04 is 2000­03. Vietnam: 1985­89 is 1986­89. b. See table 1.11, note b. c. Assumes that the rates of assistance in the Asian economies not under study are the same as the average for the focus economies and that the share of the former economies in the value of Asian agricultural production at undistorted prices is the same as the average share of the former economies in the region's agricultural GDP at distorted prices, which was 6 percent. (It was actually only 4 percent in 2004, but was somewhat larger in earlier decades.) d. Including assistance for nontradables and non-product-specific assistance. e. Including product-specific input subsidies and non-product-specific assistance. 43 44 Distortions to Agricultural Incentives in Asia Which products contribute to the positive or negative assistance to farmers because of agricultural price and trade policies? Panel c in table 1.15 shows that the negative contribution arises mainly from policies directly affecting the exportable parts of agriculture, while most of the positive contribution arises from the protection for import-competing producers. The product NRAs, shown in table 1.16, panel a, offer insights into the products that are responsible. This information is combined with information on each product's share of the gross value of production to obtain the U.S. dollar contributions by product that are shown in table 1.16, panel b (also see appendix B). Rice, milk, and sugar are clearly still the most well assisted industries. Note that, in the early 1980s, when China was still heavily taxing farmers, the net contribution of rice was negative and large. At that time, policies affecting pig meat, fruits, and vegetables were equally important contributors to the effective taxation of Asian farmers, and this was almost entirely because of the influence of China. Assistance to nonfarm sectors and RRAs The antiagricultural policy bias of past decades was not generated only by agricul- tural policies. Likewise, the significant reductions in the border protection for the manufacturing sector, which have represented the dominant intervention in the tradables part of the nonagricultural sector, have also been important in the changes in the incentives affecting intersectorally mobile resources. Reductions in the assistance for producers of nonfarm tradables have been even more responsi- ble for the improvement in farmer incentives than the reductions in the direct tax- ation on agricultural industries. It has not been possible to quantify the distortions in nonfarm tradables as carefully as we have done in agriculture. Authors have typically been obliged to rely on the applied trade taxes for exports and imports rather than price compar- isons, and they therefore usually do not capture the quantitative restrictions on trade that were once important, but have been decreasingly so through recent times.11 Nor do they capture distortions in services, and services are also now gen- erating tradables (or would be generating them in the absence of interventions preventing the emergence of service tradables). As a result, the NRA estimates for nonfarm importables are smaller and decline less rapidly than the actual NRAs. The same is true of the estimated NRAs for nonfarm exportables, except that the NRAs would have sometimes been negative. Of these two elements of underesti- mation, the former bias certainly dominates. Thus, the author estimates of the overall NRAs for nonagricultural tradables should be considered lower-bound estimates, especially in the past, and the declines are therefore less rapid than they should be.12 Table 1.16. NRA and Gross Subsidy Equivalents of Farmer Assistance, by Product, Asian Focus Economies, 1955­2004 a. NRA (at primary product level, %) Product 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Barley 41 84 72 120 101 166 357 524 543 563 Beef 38 25 34 44 95 101 94 145 106 85 Cassava -- -- -- 23 1 9 17 11 14 10 Chickpeas -- 50 24 1 0 8 12 9 15 19 Cocoa -- -- 2 3 2 2 1 2 2 0 Coconuts 29 29 24 8 3 11 19 34 22 8 Coffee -- -- -- 7 4 9 5 5 1 2 Cotton -- 19 12 63 7 12 2 3 0 5 Eggs 25 21 19 0 6 10 22 27 23 51 Fruits and vegetablesa -- 0 0 0 0 8 3 11 6 4 Jute -- -- -- 30 37 29 35 38 6 39 Maize -- 10 50 19 8 20 6 15 8 13 Milk -- -- 71 122 139 108 124 40 23 32 Oilseedsb -- 24 31 11 5 22 35 21 22 22 Palm oil -- 11 11 15 14 1 2 2 9 3 Pig meat 10 16 59 51 47 41 39 3 7 4 Poultry 25 0 69 18 58 48 2 20 17 12 Rice 10 6 25 17 13 27 6 9 2 18 Rubber 16 16 14 8 19 19 14 16 5 4 Sorghum -- 82 42 55 12 7 36 7 21 16 Sugar -- 96 163 13 2 37 39 13 20 43 Tea 22 39 39 28 22 18 19 10 8 7 Wheat 33 12 24 15 3 3 12 4 18 11 45 Totalc 11.1 2.7 2.6 0.3 0.4 21.6 15.2 4.8 6.0 10.2 (Table continues on the following page.) Table 1.16. NRA and Gross Subsidy Equivalents of Farmer Assistance, by Product, Asian Focus 46 Economies, 1955­2004 (continued) b. Gross subsidy equivalent (at undistorted farmgate prices, constant US$, millions) Product 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Barley 96 307 439 588 546 436 444 331 198 140 Beef 21 49 107 162 670 893 822 1,410 1,236 831 Cassava -- -- -- 86 7 87 136 84 80 41 Chickpea -- 1201 499 40 0 193 268 149 224 255 Cocoa -- -- 0 0 1 1 4 2 1 0 Coconut 135 104 110 543 256 841 841 1,103 1,017 273 Coffee -- -- -- 18 42 56 34 28 14 11 Cotton -- 72 338 1,820 302 1,008 147 227 20 197 Egg 28 29 33 13 41 70 150 213 199 282 Fruits and vegetablesa -- 0 0 5 19 23,014 45,349 14,769 1,304 1,239 Jute -- -- -- 338 275 147 193 118 12 66 Maize -- 36 991 620 230 4,041 977 2,530 1,336 1,926 Milk -- -- 605 756 9,163 9,044 10,865 5,162 4,405 6,459 Oilseedsb -- 1,212 1,150 455 260 1,582 2,609 2,015 2,176 1,743 Palm oil -- 13 20 210 380 70 101 63 680 210 Pig meat 62 136 668 897 1,464 35,274 16,203 1,232 3,443 2,019 Poultry 26 10 217 78 831 1,603 354 1,700 2,806 2,023 Rice 353 299 8,741 13,809 10,843 32,220 5,945 7,328 2,150 11,789 Rubber 51 438 322 356 1,389 1,230 734 813 121 48 Sorghum -- 2,036 1,017 1,240 359 186 568 30 231 125 Sugar -- 476 2,875 36 395 3,665 3,186 1,605 1,969 3,738 Tea 160 210 158 125 140 112 100 53 42 40 Wheat 19 243 1,732 928 443 513 2,457 719 4,593 2,322 Totald 704 771 1,377 7,690 902 76,336 50,000 14,889 21,676 32,010 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: -- no data are available. a. Fruits and vegetables include fruit and vegetable aggregates for China and India and bananas, cabbages, chilies, garlic, onions, peppers, and potatoes for other economies. b. Oilseeds include groundnuts, rapeseeds, soybeans, and sunflower seeds. c. For covered products only. The weights are the value of production at undistorted prices. d. For covered products only, hence, less than the totals in table 1.15, panel a. Introduction and Summary 47 Despite these methodological limitations, the estimated NRAs for nonfarm tradables were sizable prior to the 1990s. For Asia as a whole, the average NRA value has steadily declined throughout the past four or five decades as policy reforms have spread. This has therefore contributed to a decline in the estimated negative RRAs for farmers. The weighted average RRAs were below 50 percent up to the early 1970s, but improved to an average of 32 percent in the 1980s, 9 percent in the 1990s, and, now positive, were averaging 7 percent in 2000­04 (or 15 percent if the average is unweighted). The five-decade trends in the RRAs and the two component sets of NRAs for each economy are summarized in table 1.17. It is clear from figure 1.6 that the contribution of the falling positive NRAs among nonfarm producers to the rise in RRAs is greater than the corresponding contri- bution of the gradual disappearance of the negative NRAs among farmers. Has the location of the production of farm products within Asia become more efficient or less efficient as a result of policy changes over the past five decades? A set of global computable general equilibrium models relying on a time series of databases is required to answer this question properly. In the absence of such models, one crude method of addressing the question involves examining the standard deviation in RRAs across the economies of the region over time. This suggests that distortions have become more dispersed over time. The dispersion averaged 35 percent in 1960­74, 50 percent in 1975­89, and 55 percent in 1990­2004 (table 1.17, bottom row). Among the striking changes in the RRAs in individual economies in the past two decades is the shift from negative to positive RRAs in China and India (fig- ure 1.7). This is significant for the region and, indeed, for the world. The extent of the decline in nonagricultural NRAs since the early 1980s is similar in these two key countries. However, the agricultural NRAs in the two economies have differed. In China, the five-year averages have risen steadily from 45 to 6 percent. In India, they have been close to zero except for a rise in the present decade and an upward spike when international food prices collapsed in the mid-1980s (figure 1.8). This dramatic rise in the RRAs in the world's two most populous economies is significant for those examining the causes of the recent increases in international food prices. One of the contributors to the increases is said to be the growing appetite for food imports in these two economies as they industrialize and per capita incomes grow. Yet, as table 1.10 shows, both economies have remained close to self-sufficient in agricultural products over the past four decades. Undoubtedly, the steady rise in the RRAs of these two economies has supported this outcome. The rise in the RRAs may have also helped ensure that the trend in the ratio of urban to rural mean incomes, adjusted for differences in the cost of living, has been flat in China since 1980 (Ravallion and Chen 2007). Meanwhile, the rise in the RRAs in India helped ensure that the Gini coefficient hardly changed between (Text continues on page 52.) 48 Table 1.17. RRAs in Agriculture, Asian Focus Economies, 1955­2004 (percent) Economy, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of NRA, agriculture 3.3 4.9 16.3 46.1 71.8 118.6 159.8 197.6 164.8 171.9 NRA, nonagriculture 45.6 37.1 22.3 11.4 11.7 6.8 5.7 3.3 2.3 1.7 RRAa 32.6 21.4 4.8 30.5 53.9 104.8 145.9 188.2 158.8 167.3 Taiwan, Chinab NRA, agriculture 15.8 4.7 3.9 12.0 8.9 18.7 33.8 46.3 54.9 70.9 NRA, nonagriculture 8.8 9.3 8.8 7.5 7.0 5.2 4.5 2.6 1.8 1.0 RRAa 22.5 4.2 4.5 4.2 1.7 12.9 28.0 42.5 52.2 69.0 Chinac NRA, agriculture 45.2 45.2 45.2 45.2 45.2 45.2 35.5 14.3 6.6 5.9 NRA, nonagriculture 41.6 41.6 41.6 41.6 41.6 41.6 28.3 24.9 9.9 5.0 RRAa 60.5 60.5 60.5 60.5 60.5 60.5 49.9 31.1 3.0 0.9 Northeast Asia NRA, agriculture 43.1 42.5 42.2 41.3 40.0 18.4 26.2 1.7 14.7 12.0 NRA, nonagriculture 40.9 40.8 40.0 39.7 39.4 71.1 18.8 15.0 6.8 3.3 RRAa 58.2 57.7 56.6 55.7 53.7 51.9 38.0 14.2 7.4 8.5 Indonesia NRA, agriculture -- -- -- 3.8 10.4 10.5 1.9 7.5 9.7 13.9 NRA, nonagriculture -- -- -- 27.7 27.7 27.7 26.5 17.6 10.6 8.1 RRAa -- -- -- 24.7 13.6 13.5 22.5 21.3 18.3 5.4 Malaysia NRA, agriculture -- 7.6 7.9 9.4 13.7 4.9 1.4 2.6 0.2 1.5 NRA, nonagriculture -- 7.4 7.0 7.1 6.5 5.2 3.9 2.8 2.0 0.9 RRAa -- 14.0 13.9 15.5 18.9 9.6 2.4 0.3 2.2 0.6 Philippines NRA, agriculture -- 1.7 14.3 6.0 7.2 4.0 15.8 16.7 35.7 23.5 NRA, nonagriculture -- 19.0 20.3 16.3 16.3 12.9 11.0 9.9 8.6 6.4 RRAa -- 17.4 5.0 19.8 20.3 14.9 4.3 6.1 24.9 15.9 Thailand NRA, agriculture -- -- -- 23.1 15.9 2.3 6.9 6.4 1.8 0.2 NRA, nonagriculture -- -- -- 16.1 16.0 14.2 11.1 10.0 8.9 7.8 RRAa -- -- -- 33.7 27.5 14.4 16.3 14.9 6.5 7.4 Vietnamb NRA, agriculture -- -- -- -- -- -- 15.9 26.4 0.0 20.7 NRA, nonagriculture -- -- -- -- -- -- 4.3 11.2 1.5 20.8 RRAa -- -- -- -- -- -- 19.2 17.4 1.3 0.0 Southeast Asia NRA, agriculture -- 5.8 5.6 10.2 0.1 4.9 0.9 4.7 0.0 12.1 NRA, nonagriculture -- 11.5 15.4 20.2 22.0 21.1 18.0 11.5 8.2 8.1 RRAa -- 15.5 8.5 25.3 18.0 13.4 16.1 14.5 7.7 3.7 Bangladesh NRA, agriculture -- -- -- -- 3.1 3.9 17.5 2.4 8.0 4.0 NRA, nonagriculture -- -- -- -- 28.4 22.4 28.5 33.3 29.0 23.4 RRAa -- -- -- -- 19.7 21.5 8.6 26.7 28.6 15.8 Indiac NRA, agriculture 5.2 5.2 5.2 12.6 7.4 4.1 67.5 2.0 2.3 15.4 NRA, nonagriculture 113.0 113.0 113.0 83.1 64.8 59.3 48.6 15.9 12.6 5.2 RRAa 56.3 56.3 56.3 38.3 43.8 33.5 11.7 12.1 12.9 12.5 Pakistanb NRA, agriculture -- 1.0 21.7 9.3 11.8 9.3 5.9 10.2 2.6 1.5 NRA, nonagriculture -- 169.7 224.5 146.7 44.0 48.3 45.1 39.3 27.0 14.6 RRAa -- 63.8 62.4 55.9 38.6 38.6 35.1 35.2 23.0 11.5 Sri Lanka NRA, agriculture 2.7 25.7 27.6 18.5 29.0 15.4 11.2 1.3 14.0 10.8 NRA, nonagriculture 104.9 124.6 138.4 70.7 52.9 57.1 59.0 47.1 36.4 22.9 RRAa 52.5 66.6 68.0 51.6 53.5 46.2 44.3 32.9 16.3 9.8 (Table continues on the following page.) 49 50 Table 1.17. RRAs in Agriculture, Asian Focus Economies, 1955­2004 (continued) Economy, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 South Asia NRA, agriculture 4.7 4.1 4.4 9.7 7.7 1.8 47.1 0.2 2.4 12.7 NRA, nonagriculture 112.7 114.4 135.9 81.7 57.8 54.6 39.9 18.6 15.0 10.1 RRAa 50.7 51.5 56.0 39.8 41.6 33.3 5.1 15.5 14.9 3.4 Asia, unweighted averaged NRA, agriculture 12.1 8.3 2.5 3.1 3.6 5.9 18.0 15.5 19.6 25.2 NRA, nonagriculture 62.8 65.2 72.0 42.8 28.8 22.1 19.3 12.3 10.0 8.4 RRAa 44.7 38.0 34.4 27.0 22.2 13.2 1.2 2.9 8.7 15.5 Asia, weighted averagee NRA, agriculture 29.0 27.7 26.9 24.3 31.3 18.8 11.2 2.6 7.5 11.7 NRA, nonagriculture 66.8 67.1 70.9 50.3 50.3 38.3 15.4 14.9 9.6 4.3 RRAa 57.5 56.4 55.3 47.9 44.7 40.8 22.8 15.2 1.9 7.1 Dispersion, national RRAsf 21.9 30.7 36.2 37.6 41.5 51.9 56.0 65.1 50.5 50.8 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: -- no data are available. a. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. b. Pakistan: 1960­64 is 1961­64. Taiwan, China: 2000­04 is 2000­03. Vietnam: 1985­89 is 1986­1989. c. See table 1.11, note b. d. Simple averages of the weighted national average RRAs. e. Weighted averages of the national average RRAs. The weights are based on the gross value of national agricultural production at undistorted prices. f. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the national NRAs. Introduction and Summary 51 Figure 1.6. NRAs for Agricultural and Nonagricultural Tradables and the RRA, Asian Focus Economies, 1955­2004 a. Unweighted averages 80 60 40 20 percent 0 20 40 60 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year NRA, agricultural tradables NRA, nonagricultural tradables RRA b. Weighted averages 80 40 0 percent 40 80 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year NRA, agricultural tradables NRA, nonagricultural tradables RRA Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: For the definition of the RRA, see table 1.17, note a. 52 Distortions to Agricultural Incentives in Asia Figure 1.7. RRAs in Agriculture, Asian Focus Economies, 1980­84 and 2000­04 170 120 70 % RRA, 20 0 30 80 of India Rep. China China Lanka Indonesia MalaysiaVietnamThailandSri Pakistan average Bangladesh Korea, Taiwan, unweightedPhilippines Asia, economy 1980­84 2000­04 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: For the definition of the RRA, see table 1.17, note a. There are no estimates for Vietnam in 1980­84. 1984 and 2004 (table 1.1). A major issue is: Will their RRAs remain at the current neutral level of close to zero (in which case, the self-sufficiency of these economies in farm products may begin to fall)? Or will their RRAs continue to rise in the same way as they have risen in Korea and Taiwan, China and, before them, Japan and Western Europe? (We return to this issue at the end of the chapter.) Comparisons with other regions The recent regional upward shift in agricultural NRAs and RRAs toward zero and even to positive numbers is not unique to Asia. Figure 1.9 shows that similar trends, albeit less steep, have resulted from policy reforms in other developing- country regions over the past four decades. This suggests that similar political Introduction and Summary 53 Figure 1.8. NRAs and RRAs, China and India, 1965­2004 a. India 120 80 40 percent 0 40 80 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year NRA, agriculture NRA, nonagriculture RRA b. China 120 80 40 percent 0 40 80 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year NRA, agriculture NRA, nonagriculture RRA Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. 54 Distortions to Agricultural Incentives in Asia Figure 1.9. NRAs and RRAs, Africa, Asia, and Latin America, 1965­2004 a. NRAs 20 10 0 percent 10 20 30 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year Latin America and the Caribbean Africa Asia b. RRAs 10 0 10 30 percent 50 70 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 year Latin America and the Caribbean Africa Asia Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: The five-year averages are weighted. The weights are based on the gross value of total agricultural production at undistorted prices. The estimates for China before 1981 and India before 1965 are based on the assumption that the NRAs were the same as the average NRAs for those economies in 1981­84 and 1965­69, respectively, and that the gross value of production of those economies in the missing years is the same as the average share of the value of the production of the economies in total regional production in 1981­84 and 1965­69, respectively. Introduction and Summary 55 economy trends may be at work as economies develop, although farm-nonfarm household income inequality is different in Asia relative to the rest of the world (figure 1.10). In the past, it has been found that agricultural NRAs and RRAs are positively correlated with per capita incomes and agricultural comparative disad- vantage (Anderson 1995). A glance at table 1.18 suggests that Asian economies have been contributors to the trend. This is confirmed statistically in the simple regressions with country fixed effects shown in figures 1.11 and 1.12 and the mul- tiple regressions with country and time fixed effects shown in table 1.19. The CTEs of agricultural policies The extent to which farm policies have impacts on retail consumer prices for food and on prices for livestock feedstuffs depends on a wide range of elements, includ- ing the degree of processing undertaken and the extent of competition along the value chain. Like the OECD (2007), we therefore attempt only to ask about the impacts of policies on the prices buyers pay at the point on the value chain where the farm product is first traded internationally and, hence, where comparisons are made between domestic and international prices (for example, as milled rice, or Figure 1.10. Income Distribution, Asian Subregions and the World, 2000 a. East Asia 0.5 0.4 0.3 density 0.2 0.1 0 1 3 5 7 9 monthly household per capita income, 1993 purchasing power parity, log all households farm households nonfarm households (Figure continues on the following page.) 56 Distortions to Agricultural Incentives in Asia Figure 1.10. (continued) b. South Asia 0.8 0.6 0.4 density 0.2 0 1 3 5 7 9 monthly household per capita income, 1993 purchasing power parity, log all households farm households nonfarm households c. World 0.4 0.3 0.2 density 0.1 0 2 4 6 8 monthly household per capita income, 1993 purchasing power parity, log all households farm households nonfarm households Source: Bussolo, de Hoyes, and Medledev, forthcoming. Note: The vertical line is the US$1-a-day poverty line in 1993 purchasing power parity terms. Introduction and Summary 57 Table 1.18. Relative Per Capita Income, Agricultural Comparative Advantage, and NRAs and RRAs for Agricultural Tradables, Asian Focus Economies, 2000­04 Relative per Agricultural Economy capita incomea comparative advantageb NRA, % RRA, % Korea, Rep. of 212 26 137.3 167.3 Taiwan, China 232 28 61.3 69.0 China 21 58 5.9 0.9 Indonesia 17 173 12.0 5.4 Malaysia 74 107 1.2 0.6 Philippines 18 67 22.0 15.9 Thailand 38 204 0.2 7.4 Vietnam 8 301 21.2 0.0 Bangladesh 7 93 2.7 15.8 India 9 143 15.8 12.5 Pakistan 10 137 1.2 11.5 Sri Lanka 16 254 9.5 9.8 Totalc 20 80 12.0 7.1 Sources: Author compilation based on Sandri, Valenzuela, and Anderson 2007; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. a. Income per capita relative to the world average, 2000­04. World 100. b. The ratio of the share of agriculture and food in national exports to the share of agriculture and food in worldwide exports, 2000­04. c. The unweighted averages for relative per capita income and agricultural comparative advantage. The weighted averages for the NRAs and RRAs. raw sugar, or beef). If they are not supplied by national sources, we have obtained consumption data directly from the food balance sheets of the Food and Agricul- ture Organization or, in the case of minor products, indirectly by using data of the Food and Agriculture Organization on the value of trade (see FAOSTAT Database 2008) and assuming that the undistorted value of consumption is production val- ued at undistorted prices, plus imports, minus exports. We used this information to construct weights so that we could sum across commodities and countries. If there were no farm input distortions and no domestic output price distor- tions so that NRAs were entirely the result of border measures such as import or export taxes or other restrictions and if there were no domestic consumption taxes or subsidies, then the CTEs would equal the NRAs for each covered product. However, such domestic distortions exist in several Asian economies. In Korea, for example, producer prices have been well above consumer prices for several impor- tant crop products. In China, the opposite was true at least until the early 1990s. In Figure 1.11. Regressions of Real GDP per Capita and Agricultural NRAs and RRAs, Asian Focus Economies, 1955­2005 a. Regression of ln real GDP per capita on NRAs, with country fixed effects 0.6 0.3 NRA 0 0.3 0.6 1 0 1 2 3 ln real GDP per capita Asia, NRA observations Asia, fitted values coefficient standard error R2 0.17 0.01 0.19 b. Regression of ln real GDP per capita on RRAs, with country fixed effects 2 1 RRA 0 1 1 0 1 2 3 ln real GDP per capita Asia, RRA observations Asia, fitted values coefficient standard error R2 0.27 0.01 0.46 Sources: Author compilation based on Sandri, Valenzuela, and Anderson 2007; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: The dependent variable for the regressions is the NRAs or RRAs by country and year expressed as a fraction. The results are ordinary least squares estimates. The explanatory variable is the natural log of real GDP per capita expressed in US$10,000s. 58 Introduction and Summary 59 China, the producers of food staples were taxed more than the consumers were subsidized even if we take into account the in-kind partial wage payment received by many urban workers (the iron rice bowl).13 Also, because of international trade, the weights one uses to aggregate product distortion rates on the consump- tion side of the market differ from the weights one uses on the production side. Hence, aggregate CTEs differ somewhat from aggregate NRAs in each economy. This may be seen by comparing the CTEs in panel a, table 1.20 with the NRAs in table 1.11. The weighted average CTEs in the region were negative until the early 1990s, but were above zero thereafter and have increased in recent years. The vari- ance in CTEs across products is even greater now than it was before the reforms of the past two decades (see table 1.20, panel b, including the bottom row). In proportional and in per capita terms, the current transfers from consumers are clearly largest in Korea and Taiwan, China (panel a, table 1.20 and panel b, Figure 1.12. Regressions of Real Comparative Advantage and Agricul- tural NRAs and RRAs, Asian Focus Economies, 1960­2004 a. Regression of revealed comparative advantage on NRAs, with country fixed effects 1.8 1.2 0.6 NRA 0 0.6 0 1 2 3 4 revealed comparative advantage Asia, NRA observations Asia, fitted values coefficient standard error R2 0.12 0.02 0.08 (Figure continues on the following page.) 60 Distortions to Agricultural Incentives in Asia Figure 1.12. (continued) b. Regression of revealed comparative advantage on RRAs, with country fixed effects 2.1 1.5 0.9 RRA 0.3 0 0.3 0.9 0 1 2 3 4 revealed comparative advantage Asia, RRA observations Asia, fitted values coefficient standard error R2 0.26 0.03 0.09 Sources: Author compilation based on Sandri, Valenzuela, and Anderson 2007; Anderson and Valenzuela, forthcoming; estimates reported in chapters 2­12 of this volume. Note: The dependent variable for the regressions is the NRAs or RRAs by country and year expressed as a fraction. The results are ordinary least squares estimates. The explanatory variable is the revealed comparative advantage, which is the ratio of the share of agriculture and processed food in national exports to the share of agriculture and processed food in worldwide merchandise exports. The revealed comparative advantage is expressed in five-year averages. table 1.21), but, in constant (2000) U.S. dollar terms, they are also large in China, India, and Indonesia (table 1.21, panel a). In the present decade, the average trans- fers from consumers to producers in the region have amounted to around US$35 billion per year. In the early 1980s, in contrast, the transfers were from pro- ducers to consumers, and they amounted to about US$50 billion per year at the producer level for the products covered in our project. Among the covered prod- ucts, the biggest transfers are related to milk, rice, and sugar (table 1.21, panel c). The role of agricultural policies in stabilizing domestic prices An often-stated objective of food policies in Asia and elsewhere is to reduce fluc- tuations in domestic food prices and in the quantities available for consumption. (Text continues on page 69.) Table 1.19. Regressions of NRAs and Selected Determinants, Asian Focus Economies, 1960­2004 Explanatory variable (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) Ln GDP per capita 0.28* 0.21* 0.23* 0.22* 0.11 0.06 0.14 0.16* 0.38* 0.28* 0.44* 0.38* ( 0.03) ( 0.03) ( 0.03) ( 0.03) ( 0.05) ( 0.05) ( 0.06) ( 0.06) ( 0.10) ( 0.9) ( 0.10) ( 0.11) Ln GDP per capita 0.23* 0.20* 0.21* 0.21* 0.19* 0.15* 0.21* 0.18* 0.23* 0.19* 0.22* 0.21* squared ( 0.02) ( 0.01) ( 0.01) ( 0.01) ( 0.02) ( 0.02) ( 0.03) ( 0.02) ( 0.03) ( 0.02) ( 0.03) ( 0.03) Import-competing 0.33* 0.34* 0.32* 0.40* 0.41* 0.40* 0.39* 0.39* 0.39* products ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) Exportables 0.13 0.12 0.14 0.03 0.03 0.03 0.04 0.04 0.04 ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) ( 0.04) Revealed comparative 0.03* 0.07* 0.04 advantagea ( 0.01) ( 0.02) ( 0.03) Trade specialization 0.11* 0.13 0.03 indexb ( 0.03) ( 0.09) ( 0.10) Constant 0.14* 0.03 0.00 0.02 0.07* 0.11 0.05 0.07 0.49* 0.23* 0.19 0.08 ( 0.01) ( 0.03) ( 0.03) ( 0.04) ( 0.02) ( 0.04) ( 0.05) ( 0.07) ( 0.12) ( 0.11) ( 0.09) ( 0.10) R2 0.10 0.27 0.27 0.27 0.07 0.23 0.22 0.22 0.14 0.28 0.29 0.29 Number of observations 2,766 2,766 2,594 2,594 2,766 2,766 2,594 2,594 2,766 2,766 2,594 2,594 Country fixed effect No No No No Yes Yes Yes Yes Yes Yes Yes Yes Time fixed effect No No No No No No No No Yes Yes Yes Yes Sources: Author compilation based on Sandri, Valenzuela, and Anderson 2007; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: The dependent variable for the regressions is the NRAs by commodity and year. The results are ordinary least squares estimates. Standard errors are indicated in parentheses. The main explanatory variable is ln GDP per capita in $10,000s. a. The ratio of the share of agriculture and processed food in national exports to the share of agriculture and processed food in worldwide exports. World 1. b. Net exports as a ratio of the sum of exports and the sum of imports of agricultural and processed food products. World 1. 61 *Significance levels at 99 percent. 62 Table 1.20. CTEs for Policies Assisting the Producers of Covered Agricultural Products, Asian Focus Economies, 1960­2004 (at the primary product level, %) a. Aggregate CTEs, by country Economy, indicator 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 5.4 14.5 39.7 63.9 114.3 148.5 176.4 144.9 154.1 Taiwan, Chinaa 7.7 6.9 19.0 15.2 38.4 82.7 116.5 136.8 166.5 China -- -- -- -- 40.3 37.1 14.3 3.3 2.4 Indonesia -- -- 9.0 6.4 8.4 4.3 6.7 11.2 17.3 Malaysia 12.0 1.4 3.6 18.1 18.1 28.8 15.7 2.8 5.6 Philippines 5.5 12.0 4.5 7.4 3.1 23.7 22.3 40.2 26.0 Thailand -- -- 27.3 19.6 5.7 6.1 6.8 3.1 2.0 Vietnama -- -- -- -- -- 11.5 24.3 1.0 20.8 Bangladesh -- -- -- 3.1 4.6 17.0 2.9 9.3 1.9 India -- 19.1 21.8 10.8 0.6 14.1 8.7 7.4 5.3 Pakistana 0.2 28.3 9.0 17.6 11.8 6.2 13.3 5.6 1.6 Sri Lanka 5.7 14.7 2.6 19.8 11.5 6.1 3.5 20.4 14.5 Unweighted average 2.3 4.2 0.7 3.1 9.2 20.3 21.5 26.6 34.8 Weighted averageb 0.1 12.3 14.9 2.2 15.5 13.9 3.1 5.4 10.2 Dispersion, national CTEsc 7.7 24.4 26.3 26.1 47.6 61.4 66.3 59.5 64.6 b. Regional CTEs, by product Product 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Barley 65 97 57 120 326 411 341 327 Beef 33 43 98 106 95 156 106 105 Cassava -- 22 1 8 13 8 12 9 Chickpeas 5 0 0 3 5 0 3 6 Cocoa 3 4 1 2 2 3 3 0 Coconuts 24 9 3 12 22 36 25 10 Coffee -- 17 11 18 13 8 3 1 Cotton 5 14 4 16 11 15 6 4 Eggs 19 0 6 10 22 27 23 51 Fruit and vegetablesd 0 0 0 26 45 17 2 1 Jute -- 30 37 30 36 39 7 42 Maize 23 1 6 4 17 7 8 14 Milk 71 122 139 108 123 39 23 31 Oilseedse 5 5 9 18 32 19 22 23 Palm oil 1 9 3 10 1 12 15 5 Pig meat 58 49 47 37 38 4 7 4 Poultry 69 18 58 49 2 21 17 12 Rice 41 37 17 18 8 9 0 16 Rubber 52 6 19 23 19 11 2 1 Sorghum 9 30 7 0 23 5 10 1 (Table continues on the following page.) 63 64 Table 1.20. CTEs for Policies Assisting the Producers of Covered Agricultural Products, Asian Focus Economies, 1960­2004 (continued) b. Regional CTEs, by product Product 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Sugar 121 0 3 32 31 8 16 27 Tea 58 29 19 14 17 10 15 17 Wheat 8 1 7 5 21 14 8 2 Weighted averageb 12 15 2 15 14 3 5 10 Dispersion, regional product CTEsf 53 45 44 52 85 96 76 73 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: The figure is based on the assumption that the CTE is the same as the NRA derived from trade measures (that is, not including any input taxes or subsidies or any domestic producer price subsidies or taxes), except for rice, barley, wheat, and sorghum in Korea and wheat in Taiwan, China. -- no data are available. a. Pakistan: 1960­64 is 1961­64. Taiwan, China: 2000­04 is 2000­03. Vietnam: 1985­89 is 1986­89. b. The weights are consumption valued at undistorted prices, where consumption (from the FAOSTAT Database) is production, plus imports, net of exports, plus changes in the stocks of the covered products. c. A simple five-year average of the annual standard deviation around the weighted mean of the average national CTEs. d. Fruits and vegetables include fruit and vegetable aggregates for China and India and bananas, cabbages, chilies, garlic, onions, peppers, and potatoes for other economies. e. Oilseeds include groundnuts, rapeseeds, soybeans, and sunflower seeds. f. A simple five-year average of the annual standard deviation around the weighted mean of the average regional CTEs for the covered products indicated. Table 1.21. Value of CTEs for Policies Assisting the Producers of Covered Agricultural Products, Asian Focus Economies, 1965­2004 (at the primary product level, using the U.S. GDP deflator, constant 2000 US$, millions) a. Aggregate CTEs Economy 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 677 1,690 6,070 9,710 12,300 17,900 15,500 14,500 Taiwan, Chinaa 149 417 638 1,390 2,400 3,660 4,290 2,380 China -- -- -- 57,500 68,600 22,800 6,240 4,090 Indonesia -- 1,330 1,760 2,360 1160 1720 3,280 3,490 Malaysia 5 2 163 196 207 170 43 65 Philippines 467 890 607 318 1,230 1,630 3,760 1,910 Thailand -- 1,550 1,250 347 229 349 168 71 Vietnama -- -- -- -- 300 1040 5 668 Bangladesh -- 334 527 744 948 167 632 92 India 13,300 22,000 9,540 1,850 8,800 6,440 7,280 3,790 Pakistana 1,390 1,100 1,380 960 410 989 546 83 Sri Lanka 111 50 208 131 65 35 190 108 Total 10,723 25,144 3,737 48,194 44,870 10,110 18,358 31,267 All Asian economiesb 11,407 26,749 3,976 51,270 47,734 10,755 19,530 33,263 (Table continues on the following page.) 65 66 Table 1.21. Value of CTEs for Policies Assisting the Producers of Covered Agricultural Products, Asian Focus Economies, 1965­2004 (continued) b. CTEs per capita Economy 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of 22.4 51.9 164.6 247.7 294.9 408.5 339.5 305.5 Taiwan, Chinaa 11.1 27.3 37.9 75.4 121.9 176.9 202.4 105.1 China -- -- -- 69.3 61.5 19.4 5.0 3.1 Indonesia -- 10.3 13.0 14.8 6.3 9.2 16.3 16.1 Malaysia 0.7 0.4 12.6 13.3 12.7 9.0 2.1 2.7 Philippines 14.0 21.6 13.9 6.7 21.6 25.2 51.7 24.6 Thailand -- 39.7 29.1 7.4 4.4 6.2 2.9 1.1 Vietnama -- -- -- -- 6.0 15.3 0.1 8.5 Bangladesh -- 22.8 8.0 8.8 9.5 1.4 4.9 0.7 India 25.6 37.3 14.7 2.9 11.3 7.3 7.8 3.6 Pakistana 24.0 14.9 19.0 11.6 3.9 8.7 4.3 0.6 Sri Lanka 5.9 3.7 14.8 8.9 3.9 2.0 10.5 5.8 Total 6.4 13.1 1.7 20.8 17.6 3.6 6.1 9.8 All Asian economiesb 6.8 13.9 1.8 22.1 18.7 3.9 6.5 10.4 c. Regional CTEs, by product Product 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Barley 430 577 442 376 440 405 276 225 Beef 112 164 982 1,430 1,110 2,980 2,260 2,120 Cassava 0 8 0 5 7 5 6 3 Chickpeas 150 0 0 91 121 0 58 100 Cocoa 0 0 0 0 0 0 0 0 Coconuts 76 564 239 878 879 1,120 1,087 322 Coffee 0 23 22 23 15 10 12 3 Cotton 211 564 165 1,289 643 1,045 477 232 Eggs 34 13 41 70 150 215 200 284 Fruits and vegetablesc 0 8 1 18,700 45,800 15,100 1,252 1,230 Jute 0 47 189 92 146 86 9 48 Maize 578 118 186 1,140 1,500 837 1,690 2,037 Milk 608 759 9,183 9,099 10,904 5,550 4,422 6,470 Oilseedsd 211 446 674 1,748 3,003 2,380 2,945 3,150 Palm oil 0 8 9 50 17 98 255 70 Pig meat 636 839 1,418 28,008 14,856 1,510 3,270 2,140 Poultry 217 79 830 1,630 313 1,810 2,840 2,100 Rice 16,500 26,199 13,700 17,600 7,020 7,680 628 9,670 Rubber 3 3 49 72 77 87 21 10 (Table continues on the following page.) 67 68 Table 1.21. Value of CTEs for Policies Assisting the Producers of Covered Agricultural Products, Asian Focus Economies, 1965­2004 (continued) c. Regional CTEs, by product Product 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Sorghum 343 854 230 8 379 118 99 13 Sugar 2,173 835 913 3,294 2,470 1,029 1,565 4,240 Tea 17 24 27 16 16 10 22 22 Wheat 590 905 957 1,710 4,870 3,310 2,530 614 Total 10,729 25,168 3,712 48,216 44,892 10,137 18,354 31,263 Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: -- no data are available. a. Pakistan: 1960­64 is 1961­64. Taiwan, China: 2000­04 is 2000­03. Vietnam: 1985­89 is 1986­1989. b. Assumes that the rates of assistance for covered products in the Asian economies not under study are the same as the average for the focus economies and that the share of the former economies in the value of Asian agricultural production at undistorted prices is the same as the average share of the former economies in the region's agricultural GDP at distorted prices during 1990­2004, which was 6 percent. These dollar amounts do not include noncovered farm products, which account for almost one-third of agricultural output (see the bottom row of Table 1.11), nor any markup that might be applied along the value chain. c. Fruits and vegetables include fruit and vegetable aggregates for China and India and bananas, cabbages, chilies, garlic, onions, peppers, and potatoes for other economies. d. Oilseeds include groundnuts, rapeseeds, soybeans, and sunflower seeds. Introduction and Summary 69 Nowhere is this more obvious than in rice. Rather than using trade as a source of less expensive imports or an opportunity for export earnings, governments fre- quently use fluctuations in trade barriers in rice as a buffer against domestic or international shocks. Because Asia produces and consumes four-fifths of the world's rice (compared with about one-third of the world's wheat and maize), this market-insulating behavior of Asian policy makers means that, even by 2000­04, only 6.9 percent of global rice production was being traded internationally, com- pared with 14 and 24 percent for maize and wheat, respectively.14 International prices are thus much more volatile for rice than for these other grains. This means that the nominal rates of protection for rice would be above trend in years of low international prices for rice and below trend in years of high international prices for rice. Figure 1.13 reveals that this has been the case. Even if we average over all our focus economies in Southeast or South Asia, the negative correlation between the rice NRAs and the international prices for rice is high, at 0.59 for Southeast Asia and 0.75 for South Asia. This behavior is evident whether the NRA trend is upward or downward. A clear illustration is provided in the case of Malaysia, Figure 1.13. NRAs and International Prices, Rice, Asian Focus Economies, 1970­2005 a. South Asian focus economies (correlation coefficient 0.75) 600 30 20 500 10 US$ 0 400 10 NRA, price, 300 20 % 30 200 40 international 50 100 60 0 70 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 year international price South Asia, NRA (Figure continues on the following page.) 70 Distortions to Agricultural Incentives in Asia Figure 1.13. (Continued) b. Southeast Asian focus economies (correlation coefficient 0.59) 600 30 500 20 10 US$ 400 NRA, 0 price, 300 % 10 200 international 20 100 30 0 40 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 year international price Southeast Asia, NRA Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. where policies were reformed during a financial crisis in 1985­87; the growth in rice protection was reversed at that time (figure 1.14). This beggar-thy-neighbor dimension of each government's food policies reduces hugely the international role that trade among nations might play in bringing stability to the world's food markets. The more countries insulate their domestic markets, the more they export the volatility in these markets to the international marketplace. This creates a perceived need among other countries to do likewise. In most cases, volatility is exported through changes in import tariffs, but export taxes and export controls are also used by exporting countries. As domestic NRAs are adjusted by these means to changes in international prices, world prices change again in reaction so that even larger adjustments in domestic NRAs become desirable. This is a classic collective action problem. A multilateral agreement to desist is thus needed. This was sought through the Uruguay Round Agreement on Agriculture, which established tariff bindings and disciplines on administered domestic prices. Tariff bindings help reduce the extent of the problem by imposing limits on the range of tariff increases imple- mented in response to low prices. However, the bindings are so far beyond the applied import tariffs that the discipline facing food-importing members in years of low international prices is weak. Moreover, there is no corresponding WTO Introduction and Summary 71 Figure 1.14. NRAs for Rice, Malaysia, 1960­2004 240 200 160 y 6.3532x 12489 y 6.3075x 12681 % 120 NRA, 80 40 0 40 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 year Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. discipline on food export restrictions, which, as 2008 has starkly revealed, can be a serious problem in years of high international prices. Summary: what have we learned? A salient feature of price and trade policies in Asia since the 1960s is the spate of major economic reforms, including significant trade liberalization. Overall levels of nonagricultural protection have declined considerably, and this has improved the competitiveness of the agricultural sector in many economies, especially China and India. Other salient features include the gradual policy shift away from taxing agricultural exportables and, at the same time (in contrast to the situation in nonagricultural sectors), the rise in agricultural import protection. These features are captured in figure 1.15, which shows agricultural TBIs on the horizontal axis and the RRAs on the vertical axis. An economy with no antiagricul- tural bias (RRA 0) and no antitrade bias within the farm sector (TBI 0) would be located at the intersection of lines extending from the 0 point on each of the axes in figure 1.15 (indicated by a large ). China and the focus economies of South and Southeast Asia were to the southwest of this neutral point in 1980­84, but, by 2000­04, most had moved toward the northeast, closer to the neutral point. This indicates they had reduced the antitrade bias in agriculture. Most had also shifted up, closer to RRA 0, but some are now above rather than below zero. This indi- cates they are assisting farmers relative to producers of other tradable products, which, similar to the antiagricultural policy bias, can lead to wasted resources. 72 Distortions to Agricultural Incentives in Asia Figure 1.15. RRAs and TBIs in Agriculture, Asian Focus Economies, 1980­84 and 2000­04 a. 1980­84 1.7 1.2 Korea, Rep. of 0.7 % RRA, 0.2 Taiwan, China 0 Malaysia Indonesia India Thailand 0.3 Bangladesh Philippines Sri Lanka Pakistan China 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.1 0.2 TBI b. 2000­04 1.7 Korea, Rep. of 1.2 0.7 Taiwan, China % Philippines China RRA, 0.2 India 0 Indonesia Bangladesh Vietnam 0.3 Pakistan, Sri Lanka, Thailand Malaysia 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0.1 0.2 TBI Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. Note: Vietnam, 1980­84: no data are available. The following features of the Asian experience over the past five decades are worth highlighting by way of a summary of the key findings of our regional study. Since the 1950s, the region has shown a gradual shift away from the taxation of farmers relative to nonagricultural producers and the emergence, during the most Introduction and Summary 73 recent decade, of positive assistance, on average, among farmers. The decline in the estimated (negative) RRAs in the region--from less than 50 percent until the early 1970s to small positive rates in the past decade--has not been dissimilar to the trends in Africa and Latin America, though it has been more dramatic. Asian farmers were effectively taxed more than US$130 billion per year in the late 1970s and early 1980s (US$170 per person working in agriculture). Asian farmers now enjoy support worth more than US$60 billion per year (nearly US$60 per person employed on farms). Despite reforms, the dispersion in the NRAs and RRAs among farmers across Asian economies has increased rather than diminished. This means there is still scope for reducing the distortions in the use of resources in agriculture though more intensive international relocation in production, especially rice production. This finding also suggests there are political economy forces at work among economies that do not change greatly over time. In particular, the econometric results reported above suggest that the NRAs and RRAs in agriculture tend to rise as per capita incomes rise and to be higher in economies with less agricultural comparative advantage. The dispersion in NRAs among farmers has also increased rather than diminished within each Asian economy under study. This means there is still scope for reducing distortions in resource use within agriculture even in economies with average NRAs and RRAs in agriculture close to zero. As in other regions, the products in Asia showing the highest rates of distortion and gross subsidy equivalent values are rice, sugar, and milk. The antitrade bias in assistance rates within the farm sector remains strong. The NRAs for import-competing farm industries have increased over the decades we study, while the NRAs for agricultural exportables have become less negative. The fact that the average NRAs for import-competing and exportable agricultural industries have risen almost in parallel means that the TBIs have not fallen much from the peaks in the 1980s. This may be understandable from a political econ- omy viewpoint, but it means that resources are not being allocated efficiently within the farm sector. Because openness tends to promote economic growth (Commission on Growth and Development 2008), it also means that total factor productivity growth in agriculture is slower than it would have been if the remaining policy interventions were ended. Measures that restrict trade are still the most important instruments of farm assis- tance and taxation. Non-product-specific assistance and domestic taxes and subsi- dies on farm inputs and outputs made only minor contributions to the NRA esti- mates for the region as a whole. Input subsidies have played a significant role in India and, occasionally, in other parts of the region; but, as in Latin America, there has been comparatively little assistance provided through public investments in 74 Distortions to Agricultural Incentives in Asia rural infrastructure and agricultural research and development, even though the social rates of return from such investments are high (Fan and Hazell 2001; Lopez and Gallinato 2006; Fan 2008).15 This suggests these other instruments could play a bigger role in boosting farm output and productivity in Asia. Movements in the CTEs closely follow changes in farm taxation and support because agricultural taxation and assistance are mostly generated through trade measures. This means that, before the reforms, food prices were kept artificially low in Asia, but, in recent years, they have been above international levels, on aver- age. This also means there is considerable variation in the CTEs across products and countries in the region. The current level of taxation of food consumers is ris- ing in the region. In 2000­04, it amounted to US$11 per capita per year. This con- trasts with the subsidy of US$22 per capita per year in 1980­84. The reductions in negative RRAs have been caused by cuts in the protection for nonagricultural sectors as much as by reforms in agricultural policies. Thus, the reductions in the distortions to agricultural incentives in the region have been caused not merely by reforms in farm policies, but are part of economy-wide reform programs. Governments continue to seek to reduce fluctuations in domestic food prices and in the quantities of food available for consumption by adding to or reducing barriers to trade. This beggar-thy-neighbor dimension of each economy's food policies reduces the role that trade among nations can play in encouraging stability in the world's food markets. This is especially the case in rice. Rice is the main staple in Asia; and Asia accounts for more than 80 percent of the global market for rice. Where To from Here? The expectation is that, provided they remain open, continue to free up domestic markets, and practice good macroeconomic governance, Asia's developing economies will grow rapidly in the foreseeable future. The growth will be more rapid in manufacturing and service activities than in agriculture. In the more densely populated economies of the region, growth will be accompanied by increases in incomes among low-skilled workers wherever labor-intensive exports boom. Agricultural comparative advantage is thus likely to decline in these economies. Whether these economies become more dependent on imports of farm products depends, however, on the RRAs. The first wave of Asian industrializers (Japan, then Korea, and Taiwan, China) chose to slow the growth of food import dependence by raising the NRAs in agriculture even as they reduced the NRAs for nonfarm tradables so that the RRAs rose above the neutral zero level. A key question is: will later industrializers follow suit, given the close associations among RRAs, rising per capita incomes, and declining agricultural comparative advantage? Introduction and Summary 75 Figure 1.16. RRAs and Real Per Capita GDP, India, Japan, and Northeast Asian Focus Economies, 1955­2005 200 150 100 % 50 RRA, 0 50 100 7 8 9 10 11 In real GDP per capita India China Taiwan, China Japan Korea, Rep. of Sources: Author compilation; Anderson and Valenzuela, 2008; estimates reported in chapters 2­12 of this volume. We have mapped the RRAs for Japan, Korea, and Taiwan, China against real per capita GDP and then superimposed on the same map the RRAs for lower-income economies to understand the extent to which the latter economies are tracking the former economies. Figure 1.16 shows the results for China and India. It indicates that the RRA trends of the past three decades in these two economies are on the same trajectory as the earlier trends in the richer economies of Northeast Asia. This represents a reason to expect the governments of later industrializing economies to follow suit, all else being equal. One reason we might expect different government behavior now is because the earlier industrializers were not bound under the General Agreement on Tariffs and Trade to keep agricultural protection down. Had there been strict discipline on farm trade measures at the time Japan and Korea joined the General Agree- ment on Tariffs and Trade in 1955 and 1967, respectively, their NRAs might have plateaued at less than 20 percent (figure 1.17). At the time of China's accession to the WTO in December 2001, the average national NRAs were below 5 percent, according to our study, or 7.3 percent for import-competing agriculture alone. The average bound import tariff commitment of China was about twice this level (16 percent in 2005), but more important were China's out-of-quota bindings on the items the imports of which were restricted through tariff rate quotas. In 2005, 76 Distortions to Agricultural Incentives in Asia Figure 1.17. NRAs for China, Japan, and the Republic of Korea and GATT/WTO Accession, 1955­2005 200 Korea, Rep. of China (1967 7.4) (2001 4.5) 150 Japan (1955 16.6) 100 % 50 NRA, 0 50 100 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­05 year Japan Korea, Rep. of China Sources: Author compilation. Note: WTO accession also refers to accession to the WTO predecessor organization, the General Agreement on Tariffs and Trade. the latter tariff bindings were 65 percent for grains, 50 percent for sugar, and 40 percent for cotton (WTO, ITC, and UNCTAD 2007). The Chinese government also had bindings on product-specific domestic supports in agriculture of 8.5 per- cent and was able to provide another 8.5 percent as non-product-specific assis- tance if it so wished. This sums to an NRA of 17 percent through domestic sup- port measures alone, which is in addition to the assistance available through out-of-quota tariff protection. Clearly, the legal commitments the Chinese government made on acceding to the WTO are distant from the current levels of domestic and border support it provides to its farmers. The commitments are therefore unlikely to constrain the government much in the next decade or so, and the legal commitments made by the Asian developing economies that joined the WTO earlier (except Korea) are even less constraining (Anderson, Martin, and Valenzuela 2008). In Bangladesh, India, and Pakistan, for example, the estimated NRAs for agricultural importables in 2000­04 are 6, 34, and 4 percent, respectively, whereas the average bound tariffs on the agricultural imports in these economies were 189, 114, and 96 percent, Introduction and Summary 77 respectively (WTO, ITC, and UNCTAD 2007). Also, like other developing- economy governments, the governments of these economies have significant bindings, 10 percent, on product-specific domestic supports and another 10 per- cent for non-product-specific assistance, a total of 20 extra percentage points of NRA that legally could be applied through domestic support measures. This com- pares with the 10 percent in India and the less than 3 percent in the rest of South Asia (tables 1.13 and 1.14). One can only hope that China and South and Southeast Asia will not make use of the legal wiggle room they have allowed themselves in their WTO bindings and thereby follow Japan, Korea, and Taiwan, China into substantial agricultural pro- tection.16 A much more efficient and equitable strategy would involve treating agriculture in the same way they have been treating nonfarm tradables. This would involve opening the sector to international competition and relying on more-efficient domestic taxes (for example, taxes on income and consumption or value added taxes) rather than trade taxes to raise government revenue. It might be argued that such a laissez faire strategy might increase rural-urban inequality and poverty and thereby generate social unrest. Nonetheless, policies that lead to high prices for staple foods, in particular, involve potentially serious risks for the urban and rural poor who are net buyers of food in developing coun- tries, as has been demonstrated during the recent increases in the prices for these goods (Ivanic and Martin 2008). The available evidence suggests that rural-urban poverty gaps have been reduced in parts of Asia when more mobile members of farm households are able to find full- or part-time work off the farm and repatri- ate a portion of their higher earnings among other farm household members (Otsuka and Yamano 2006; World Bank 2007). Concerted government interven- tion through social policy measures are hugely important in reducing the gaps between rural and urban incomes and in raising national incomes overall (Winters, McCulloch, and McKay 2004; Hayami 2007). Efficient ways of assisting any poor (nonfarm or farm) households left behind include public investment measures that have high social payoffs, such as in basic education and health care, and investments in rural infrastructure and agricultural research and develop- ment.17 What do the above lessons and implications suggest developing-country policy makers should do when confronted, as in recent years, by a sharp upward move- ment in international food prices? In the past, as illustrated for rice in figures 1.13 and 1.14, many governments have simply increased export restrictions or lowered import restrictions on food staples for the duration of the spike. What if the recent rise in international prices is more prolonged than the short-lived spikes of the past? Outlook projections issued by international agencies suggest that prices might remain high for the foreseeable future and that growth in net food imports 78 Distortions to Agricultural Incentives in Asia by the rapidly industrializing economies of Asia is a significant contributor to this phenomenon.18 Yet, as we see in figure 1.8, China and India over the past two or three decades have steadily raised the RRAs, though these had been adequate to keep both countries close to self-sufficiency in primary agricultural products over the previous four decades. However, in 2000­04, China became a net importer in all trade in agriculture and processed food for the first time, while, in South Asia, the net exports of India were less than the net imports of Bangladesh and Pakistan for the first time since the late 1960s (Sandri, Valenzuela, and Anderson 2007).19 Should these governments choose to maintain RRAs at current levels (close to zero), the import dependence of these economies in agriculture might rise. If this occurs, other developing economies might reconsider their current position in the WTO's Doha Round of trade negotiations. By agreeing to lower substantially their bound tariffs and their subsidies on agricultural products, they could extract greater concessions from high-income countries without having to reduce their actual applied rates in the foreseeable future. We have dealt little in this chapter with the effects of the trends and fluctuations in the NRAs and RRAs, especially on economic welfare, price stability, income equality, or poverty, and we have dealt little with the underlying causes of these effects. The analytical narratives in the chapters that follow touch on these issues, but more in-depth empirical analysis may now be undertaken with the help of our assistance estimates. Some analysis will appear in our project's publications. Specif- ically, Anderson, Valenzuela, and van der Mensbrugghe (forthcoming) provide results from a global economy-wide model of the impacts on agricultural markets, national economic welfare, and net farm incomes of distortions to the world's goods markets as of 2004. They use the new agricultural distortion estimates in this volume and its companion volumes. How these distortions, both within an econ- omy and in the rest of the world, affect poverty and inequality are explored in a series of country case studies brought together in Anderson, Cockburn, and Martin (forthcoming), who use national economy-wide models that have been enhanced with detailed earning and spending information on numerous types of urban and rural households. In Anderson (Political Economy, forthcoming), a broad range of theoretical and econometric analyses are assembled to shed light on the political economy forces that have generated the evolving pattern of inter- and intrasectoral distortions in farmer and food consumer incentives over the past half century. Our hope is that these and others results of this study and its companion volumes will spawn much more analysis in the years to come. Notes 1. The other three regional studies are Anderson and Masters (forthcoming), Anderson and Swinnen (2008), and Anderson and Valdés (2008). Together with comparable studies of high-income countries, including Japan, they have formed the basis for a global overview volume (Anderson, A Global Perspective, forthcoming). Introduction and Summary 79 2. Apart from the urban islands of Hong Kong, China and Singapore, the region's economies that have populations above 1 million each and that we have omitted are Afghanistan, Cambodia, the Democratic People's Republic of Korea, the Lao People's Democratic Republic, Mongolia, Myanmar, Nepal, and Papua New Guinea, which, in 2004, contributed less than 5 percent of the population, 4 percent of the agricul- tural GDP, and 2 percent of the total GDP of all the developing economies in the region. 3. Anderson and Hayami (1986) only examine Japan, the Republic of Korea, and Taiwan, China and only from 1955 to 1982 (apart from the rice distortion estimates, which are provided for an addi- tional five decades). Krueger, Schiff, and Valdés (1991) analyze Korea, Malaysia, Pakistan, the Philip- pines, Sri Lanka, and Thailand, but for only four or five crops and only from 1960 to 1984. Orden et al. (2007) supply producer support estimates only for China, India, Indonesia, and Vietnam and only since 1985. The OECD (2007) has been covering its two Asian members, Japan and Korea, since 1986, but has also begun examining China (see OECD 2005). All these earlier studies find that the average nominal rates of assistance (NRAs) for farmers are higher in higher-income settings and in settings in which agricultural comparative advantage is weaker. They find that the NRAs in each economy are also much higher in the import-competing subsector than among exporters of farm products. 4. In overall land endowment per capita rather than only cropland and pastureland, China has around 30 percent of the global average. 5. The regime changes that have occurred during this period have included the shift from socialism to the market in China and Vietnam and the opening of India and several other economies. The region saw the end of colonization between the late 1940s and late 1950s. The Korean Peninsula and Taiwan, China obtained independence from Japan in 1945, India and Pakistan from Britain in 1947, Indonesia from the Netherlands in 1949, Indochina from France in 1954, and Malaya from Britain in 1957. 6. The nine are China; Hong Kong, China; Indonesia; Japan; Korea; Malaysia; Singapore; Taiwan, China; and Thailand. Brazil is the only other large economy in the set of 12; the other three economies are Botswana, Malta, and Oman. 7. By contrast, the share of our focus economies in the global exports of agricultural and food products grew only a little, from 11.0 to 13.5 percent between 1990 and 2006. 8. Our definition of a policy-induced price distortion follows Bhagwati (1971) and Corden (1997). It includes any policy measure at the border (such as a trade tax or subsidy, a quantitative restriction on trade, or a dual or multiple foreign exchange rate system, assuming the country is small enough to have no monopoly power in international markets) or any domestic producer or consumer tax, sub- sidy, or restraint on output, intermediate inputs, or primary factors of production (except where these are needed to overcome directly an externality, or where it is set optimally across all products or fac- tors, for example as a value added tax to raise government revenue). 9. Corden (1971) proposed that the volume of free trade might be used as weights, but because this is not observable (and an economy-wide model is needed to estimate the weights), the usual practice is to compromise by using actual distorted volumes, but undistorted unit values or, equivalently, dis- torted values, divided by (1 NRA). If estimates of own and cross-price elasticities of demand and supply are available, a partial equilibrium estimate of the quantity at undistorted prices might be gen- erated, but, if the estimated elasticities are unreliable, this may introduce more error to the error one seeks to correct. 10. Recall that, unless otherwise noted, Asia is shorthand throughout this volume for our focus developing economies and thus excludes Japan. 11. Also ignored are distortions in the prices for the inputs into the production of nonfarm goods, which is in contrast to the treatment of these prices in agricultural NRA estimates. 12. This bias is accentuated in cases in which distortions in exchange rates are not included, as noted also in the section on methodology. Exchange rate distortions have been included in the studies that form the basis of the chapters on China, Malaysia, Pakistan, Sri Lanka, and Vietnam. Their impact was greatest in China, where they caused the RRAs to be more negative by about 2 percentage points in the 1970s, 6 percentage points in the 1980s, and 3 percentage points in the 1990s (see chapter 3, table 3.5 in this volume). 13. It is also true that consumers in South Asia received staple foods at prices that were effectively subsidized through the fair price shops. However, these government handouts were sufficiently 80 Distortions to Agricultural Incentives in Asia rationed to be mostly inframarginal and, so, are considered here as implicit income transfers rather than additional distortive consumer subsidies. 14. The number for the share of rice production represented an increase over the pre-1990s half- decade global shares, which were all less than 4.5 percent (for example, 4.1 percent in 1985­89) and is greater than the Asian share of only 5.7 percent in 2000­04 according to our estimates. 15. Data in Pardey et al. (2006) suggest that public research and development expenditure in Asia since the late 1970s has averaged less than 0.5 percent of the gross value of production at undis- torted prices. This is trivial compared with the NRAs for the region that are generated through price- distorting measures. These NRAs are 25 to 40 times this share of the gross value of production (below 20 percent until the mid-1980s and 12 percent in 2000­04). 16. The indications in the ongoing Doha Round of multilateral trade negotiations at the WTO are not encouraging. The Group of 33 developing countries, led by Indonesia, but strongly supported by India and the Philippines, among others, is arguing for additional special and differential treatment for developing countries in the form of exemptions from agricultural tariff cuts for special products and for a special safeguard mechanism that would allow these countries to impose tariffs that are even higher than the bound tariffs in years of likely import surges. 17. As implied by the estimates reported in note 15 above, even if only 5 percent of the current NRAs provided to Asian farmers through farm price support policies were replaced by agricultural research and development expenditure, this would more than double current public spending on this research and development. The latter would increase regional economic welfare, whereas price- distortionary policies reduce welfare. Such a boost in research and development might generate another Green Revolution, especially if it took full advantage of the new developments in biotechnol- ogy. (For rice, for example, see Anderson, Jackson, and Nielsen 2005.) 18. The World Bank's commodity forecasts indicate that, by 2020, grain prices will still be 10 per- cent above 2006 levels in real terms (Prospects for the Global Economy Database 2008). The 2006 lev- els were 20 percent above the average for 2001­05. The International Food Policy Research Institute, the OECD, and the Food and Agriculture Organization similarly expect food prices to remain high well into the next decade and beyond (von Braun 2007; OECD and FAO 2008). 19. The change in China was largely caused by increases in the cotton imports needed to supply the economy's rising production in textiles and clothing for export. References Anderson, K. 1995. "Lobbying Incentives and the Pattern of Protection in Rich and Poor Countries." Economic Development and Cultural Change 43 (2): 401­23. ------. 2008. "Distorted Agricultural Incentives and Economic Development: Asia's Experience." 2008 World Economy Lecture, Leverhulme Center for Research on Globalization and Economic Policy, University of Nottingham, Semenyih, Malaysia, January 16. Forthcoming in The World Economy 32 (2009). ------, ed. forthcoming. Distortions to Agricultural Incentives: A Global Perspective, 1955 to 2007. ------, ed. forthcoming. Political Economy of Distortions to Agricultural Incentives. Anderson, K., J. Cockburn, and W. Martin, eds. forthcoming. Agricultural Price Distortions, Inequality, and Poverty. Anderson, K., and Y. Hayami, eds. 1986. The Political Economy of Agricultural Protection: East Asia in International Perspective. London: Allen and Unwin. Anderson, K., L. A. Jackson, and C. P. Nielsen. 2005. "GM Rice Adoption: Implications for Welfare and Poverty Alleviation." Journal of Economic Integration 20 (4): 771­88. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Introduction and Summary 81 Anderson, K., W. Martin, and E. Valenzuela. 2008. "Long Run Implications of WTO Accession for Agri- culture in China." In China's Agricultural Trade: Issues and Prospects, ed. C. Carter and I. Sheldon. London: CABI. Anderson, K., and W. Masters, eds. forthcoming. Distortions to Agricultural Incentives in Africa. Washington, DC: World Bank. Anderson, K., and J. Swinnen, eds. 2008. Distortions to Agricultural Incentives in Europe's Transition. Washington, DC: World Bank. Anderson, K., and A. Valdés, eds. 2008. Distortions to Agricultural Incentives in Latin America. Washington, DC: World Bank. Anderson, K., and E. Valenzuela. 2008. "Estimates of Global Distortions to Agricultural Incentives, 1955 to 2007." World Bank. http://go.worldbank.org/5XY7A7LH40. Anderson, K., E. Valenzuela, and D. van der Mensbrugghe. forthcoming. "Effects of Distortions on Global Welfare, Farm Incomes, and Agricultural Markets." In Distortions to Agricultural Incentives: A Global Perspective, 1955 to 2007, ed. K. Anderson, chap. 12. London: Palgrave Macmillan; Wash- ington, DC: World Bank. Bhagwati, J. N. 1971. "The Generalized Theory of Distortions and Welfare." In Trade, Balance of Pay- ments and Growth: Papers In International Economics in Honor of Charles P. Kindleberger, ed. J. N. Bhagwati, R. W. Jones, R. A. Mundell, and J. Vanek, 265­86. Amsterdam: North-Holland. Bussolo, M., R. de Hoyes, and D. Medledev. forthcoming. "Policy Impacts on Global Income Distribu- tion, Inequality, and Poverty Using Microsimulation with the Linkage Model." In Agricultural Price Distortions, Inequality, and Poverty, ed. K. Anderson, J. Cockburn, and W. Martin, chap. 4. Chen, S., and M. Ravallion. 2007. "Absolute Poverty Measures for the Developing World, 1981­2004." Policy Research Working Paper 4211, World Bank, Washington, DC. Commission on Growth and Development. 2008. The Growth Report: Strategies For Sustained Growth and Inclusive Development. Washington, DC: Commission on Growth and Development, World Bank. Corden, W. M. 1971. The Theory of Protection. Oxford: Clarendon Press. ------. 1997. Trade Policy and Economic Welfare. 2nd ed. Oxford: Clarendon Press. Fan, S., ed. 2008. Public Expenditures, Growth, and Poverty: Lessons from Developing Countries. Baltimore: Johns Hopkins University Press; Washington, DC: International Food Policy Research Institute. Fan, S., and P. Hazell. 2001. "Returns to Public Investments in the Less-Favored Areas of India and China." American Journal of Agricultural Economics 83 (5): 1217­22. FAO Agricultural Trade Database (FAOSTAT). Food and Agriculture Organization of the United Nations. http://faostat.fao.org/site/342/default.aspx (accessed May 2008). FAOSTAT Database. Food and Agriculture Organization of the United Nations. http://faostat.fao.org/ default.aspx (accessed May 2008). FAO SUA-FBS Database (Supply Utilization Accounts and Food Balance Sheets Database, FAOSTAT). Food and Agriculture Organization of the United Nations. http://faostat.fao.org/site/354/default .aspx (accessed May 2008). Hayami, Y. 2007. "An Emerging Agricultural Problem in High-Performing Asian Economies." Policy Research Working Paper 4312, World Bank, Washington, DC. Ivanic, M., and W. Martin. 2008. "Implications of Higher Global Food Prices for Poverty in Low- Income Countries." Policy Research Working Paper 4594, World Bank, Washington, DC. Krueger, A. O., M. Schiff, and A. Valdés. 1988. "Agricultural Incentives in Developing Countries: Meas- uring the Effect of Sectoral and Economywide Policies." World Bank Economic Review 2 (3): 255­72. ------. 1991. Asia. Vol. 2 of The Political Economy of Agricultural Pricing Policy. World Bank Compar- ative Study. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Lerner, A. P. 1936. "The Symmetry between Import and Export Taxes." Economica 3 (11): 306­13. Lloyd, P. J. 1974. "A More General Theory of Price Distortions in an Open Economy." Journal of Inter- national Economics 4 (4): 365­86. 82 Distortions to Agricultural Incentives in Asia Lopez, R., and G. I. Gallinato. 2006. "Should Governments Stop Subsidies to Public Goods? Evidence from Latin America." Journal of Public Economics 91 (5­6): 1071­94. Maddison, A. 2003. The World Economy: Historical Statistics. Paris: Development Centre, Organisation for Economic Co-operation and Development. OECD (Organisation for Economic Co-operation and Development). 2005. Review of Agricultural Policies: China. Paris: OECD. ------. 2007. Agricultural Policies in OECD Countries: Monitoring and Evaluation 2007. Paris: OECD. OECD (Organisation for Economic Co-operation and Development) and FAO (Food and Agriculture Organization of the United Nations). 2008. OECD-FAO Agricultural Outlook 2008­2017. Paris: OECD. OECD PSE-CSE Database (Producer and Consumer Support Estimates, OECD Database 1986­2005). Organisation for Economic Co-operation and Development. http://www.oecd.org/document/ 55/0,3343,en_2649_33727_36956855_1_1_1_1,00.html (accessed December 2007). Orden, D., F. Cheng, H. Nguyen, U. Grote, M. Thomas, K. Mullen, and D. Sun. 2007. Agricultural Pro- ducer Support Estimates for Developing Countries: Measurement Issues and Evidence from India, Indonesia, China, and Vietnam. Research Report 152. Washington, DC: International Food Policy Research Institute. Otsuka, K., and T. Yamano. 2006. "Introduction to the Special Issue on the Role of Nonfarm Income in Poverty Reduction: Evidence from Asia and East Africa." Agricultural Economics 35 (3) (supple- ment): 373­97. Pardey, P. G., N. Beintima, S. Dehmer, and S. Wood. 2006. "Agricultural Research: A Growing Global Divide?" Food Policy Report 17, International Food Policy Research Institute, Washington, DC. PovcalNet. World Bank. http://go.worldbank.org/7X6J3S7K90 (accessed May 2008). Prospects for the Global Economy Database. World Bank. http://go.worldbank.org/PF6VWYXS10 (accessed May 2008). Ravallion, M., and S. Chen. 2007. "China's (Uneven) Progress against Poverty." Journal of Development Economics 82 (1): 1­42. Ravallion, M., and G. Datt. 1996. "How Important to India's Poor is the Sectoral Composition of Economic Growth?" World Bank Economic Review 10 (1): 1­26. Sandri, D., E. Valenzuela, and K. Anderson. 2007. "Economic and Trade Indicators for Asia, 1960 to 2004." Agricultural Distortions Working Paper 20, World Bank, Washington, DC. von Braun, J. 2007. "The World Food Situation: New Driving Forces and Required Actions." Food Policy Report 18, International Food Policy Research Institute, Washington, DC. Vousden, N. 1990. The Economics of Trade Protection. Cambridge, UK: Cambridge University Press. Winters, L. A., N. McCulloch, and A. McKay. 2004. "Trade Liberalization and Poverty: The Empirical Evidence." Journal of Economic Literature 62 (1): 72­115. World Bank. 2007. World Development Report 2008: Agriculture for Development. Washington, DC: World Bank. World Development Indicators Database. World Bank. http://go.worldbank.org/B53SONGPA0 (accessed May 2008). WTO (World Trade Organization). International Trade Statistics 2007. Geneva: WTO. WTO (World Trade Organization), ITC (International Trade Center), and UNCTAD (United Nations Conference on Trade and Development). 2007. World Tariff Profiles 2006. Geneva: WTO. Part II II NORTHEAST ASIA 2 Republic of KOREA AND TAIWAN, CHINA Masayoshi Honma and Yujiro Hayami The story of agricultural policies over the past 50 years in the Republic of Korea (the southern part of the Korean Peninsula, hereafter referred to as Korea) and Taiwan, China includes the dramatic changes in the distortions to agricultural incentives that producers and consumers have face in the course of the develop- ment of these economies. In this study, we estimate the degree of distortions for key agricultural products, as well as for the agricultural sector as a whole, over a period when these economies transitioned from low- to high-income economies (from 1955 to 2004). Schultz (1978) established that, as economies advance from low-income to high-income status, price and trade policies tend to switch from taxing to subsi- dizing agriculture relative to other tradable sectors. During the 50 years of our analysis, Korea and Taiwan, China jumped from the status of low-income to middle-income economies and toward the status of high-income economies. Policy switching--shifting from negative to positive assistance for agricultural producers in the course of economic development--is clearly observable in the time series on these two economies. We compare the switching processes in the two economies--which grew at about the same speed--to obtain insights into the effects of policies and the underlying causes of the changes in the distortions in agricultural incentives. Our findings shed light on the process of change in agricultural distortions that occurs during the stages in economic development. In the following sections, we first briefly describe the structure of agriculture in the course of the development of these two economies. Next, the evolution of agricultural policies in these economies is reviewed. Distortions to agricultural 85 86 Distortions to Agricultural Incentives in Asia incentives are measured in terms of the nominal and relative rates of assistance to agriculture (NRAs and RRAs to agriculture, respectively). We then discuss the pol- icy implications of the estimates for the two economies and draw lessons for less- developed economies that are currently experiencing structural transformation in the course of growth. Economic Development and Structural Change The available choices among agricultural policies, particularly price-distorting policies, are closely linked to the process of economic development. As identified by Schultz (1978), there are two agricultural problems. The food problem under- lies the policies commonly adopted in low-income countries that exploit or tax agriculture. Such policies are in contrast to policies in many high-income coun- tries that protect or subsidize agriculture to solve the farm problem. The hypoth- esis of Schultz became an established paradigm among agricultural economists and found support in several empirical studies (Anderson and Hayami 1986; Hayami 1988; Krueger, Schiff, and Valdés 1991). More recently, Hayami and Godo (2004) and Hayami (2005) have added the disparity problem, which is specific to middle-income economies. They suggest that one should examine the ways dis- tortions in agricultural incentives change in all three types of economies over the stages of development. The most distinguishing characteristic of Korea and Taiwan, China during the period of analysis is the unusually rapid rates of economic growth generated by success in industrial development. Indeed, during the East Asian miracle (World Bank 1993), while Japan was first, Korea and Taiwan, China--together with Hong Kong, China and Singapore--comprised the runner-up group; and China and members of the Association of Southeast Asian Nations are now following these economies. Since economic growth is a fundamental determinant of the nature of distortions to agricultural incentives, it is useful, in this section, to offer an overview of development processes in Korea and Taiwan, China. An historical perspective Because of regular monsoon rains and the mountainous, undulated topography in which water may be controlled relatively easily through cooperation at the fam- ily and community levels, this region is well suited to rice production on small family farms organized into village communities. The agrarian structure therefore became characterized by a uni-modal distribution of smallholders who farmed an average plot of about 1 hectare. Relative to Southeast Asia, large agribusiness plan- tations based on hired labor were almost completely absent not only in Korea, Republic of Korea and Taiwan, China 87 which is located in the temperate zone, but also in Taiwan, China, where tropical cash crops such as sugar and bananas occupy a significant agricultural subsector. The rural community was traditionally stratified across landlords, landowning cultivators, and landless tenants. Agricultural laborers subsisting on hired labor wages were not a significant component of the rural population in this region. Land reforms after World War II significantly changed the distribution of landownership, but the distribution of operational landholdings was essentially intact. There is a high degree of similarity in the agrarian structures of the two economies partly because Japan brought its own institutions to its colonies: Taiwan, China was ceded by China to Japan in 1895, and the Korean Peninsula was annexed by Japan in 1910. The biggest reform during Japanese colonization was the transfer of fee simple titles to landowners through cadastral surveys in return for the commitment of the new landowners to pay the land tax. Japanese efforts to develop the colonies concentrated on agriculture, especially rice, when shortages within Japan became evident after the Rice Riots of 1918. The promo- tion of rice production through agricultural research and extension systems, as well as improvements in irrigation and drainage infrastructure, plus protection from rice imports from the rest of the world (see Anderson and Tyers 1992), was a major success from the viewpoint of the colonists in that Japan's rice imports from the two colonies rose from 5 to 20 percent of the country's consumption between 1915 and 1935. Rising exports of rice and other primary commodities and the corresponding inflows of manufactured commodities meant that dependency on agriculture remained significant in the two economies. This was especially so in the southern half of the Korean Peninsula because Japanese efforts in industrial development in Korea were concentrated in the north to exploit the hydroelectric power of the Yalu River and feed a network of chemical industries that was larger than the one in Japan. The heavy dependency on agriculture in the south was intensified by the urban destruction during the Korean War (1950­53). In Taiwan, China, com- merce and industry were more active because the relatively larger cash crop sub- sector required larger amounts of processing and marketing relative to Korean agriculture, which was dominated by subsistence crops such as rice and barley. Today, the wide dispersion of small and medium industries in rural areas in Taiwan, China contrasts with the concentration of Korean industry in large-scale establishments in urban areas and seems to be, at least in part, rooted in the cash crop tradition. We discuss elsewhere why Japan, Korea, and Taiwan, China were able to achieve remarkable success in economic development as forerunners to the East Asian miracle (see Hayami and Aoki 1998; Hayami and Godo 2005). Here, it suffices to 88 Distortions to Agricultural Incentives in Asia say that their success was the result of borrowing technology from advanced economies. Gerschenkron (1962) suggested that the later the start of industrial- ization, the larger the scope for achieving economic growth by borrowing tech- nology. Why then were Korea and Taiwan, China, in particular, so successful in this borrowing in the midst of many other late starters? One reason is that the two economies were endowed with cheap, but relatively well-educated labor because of the diffusion of compulsory elementary education systems during the colonial period. This made the initial borrowing of labor-intensive industrial technologies more rewarding in these resource-poor economies. The accumulation of human capital through education investments by the newly independent governments smoothed the way for the switch to capital- and knowledge-intensive technologies that occurred later. Another reason was the constant danger of Communist aggression, which compelled leaders to adopt policies to achieve economic success for the sake of maintaining legitimacy and viability. The success of development based on industrial technology borrowing by their neighbor Japan was an added motive for adopting this strategy. There are many similarities, but also significant differences in the industrial- ization strategies followed by Korea and Taiwan, China. In the latter, although the Nationalist Party exerted strong control over formal sectors, the government did not attempt to intervene in the activities of small and medium entrepreneurs in informal sectors. Informal sectors were able to grow by developing various marketing and financial links among themselves and with foreign firms. These enterprises were dispersed widely over urban and rural areas and were able to achieve substantial international competitive strength (Ho 1979, 1982). In contrast, government control in Korea was stronger and more complete, espe- cially under the military administration of Pak Chong-hui (1961­79). All formal credits were channeled from nationalized banks to large industries, while foreign direct investment was tightly controlled. The strategy was the reason for the high concentration of industrial production in Korea in a small number of large enterprises located predominantly in urban areas (Cole and Park 1983; Amsden 1989). Economic growth and structural transformation We now provide a quantitative summary of economic development in Korea and Taiwan, China during the past five decades. Table 2.1 shows selected indicators of economic development. The first rows provide data on real gross domestic prod- uct (GDP) per capita in 2000 constant prices in purchasing power parity dollars in Korea and Taiwan, China. Until 1960, the two economies showed a low level of per capita GDP, at less than US$1,500. Thereafter, income levels rose rapidly, Table 2.1. Economic Growth and Structural Transformation, Republic of Korea and Taiwan, China, 1955­2004 Indicator Economy 1955 1960 1970 1980 1990 2000 2004 Real GDP per capita, 2000 Korea, Rep. of 1,429 1,458 2,552 4,497 9,593 15,702 18,424 constant prices, US$ Taiwan, China 1,241 1,444 2,846 5,963 11,248 19,184 20,868 Agriculture in GDP,a % Korea, Rep. of 46.9 39.1 29.2 16.2 8.9 4.9 3.8 Taiwan, China 28.9 28.2 15.3 7.5 4.0 2.0 1.7 Agriculture among the Korea, Rep. of 79.7 60.2 49.1 37.1 18.1 10.0 7.7 economically active population,a % Taiwan, China 53.6 50.2 36.7 19.5 12.8 8.9 7.5 Farm household population Korea, Rep. of 61.9 58.2 44.7 28.4 15.5 8.6 7.1 in total population, % Taiwan, China 50.7 49.8 40.9 30.3 21.1 16.5 14.3 Agricultural GDP per worker in Korea, Rep. of 58.8 65.0 59.5 43.7 49.2 49.0 49.4 total GDP per worker, % Taiwan, China 53.9 56.2 41.7 38.5 31.3 22.5 22.7 Sources: Heston, Summers, and Aten 2006; MAF, various; Department of Agriculture and Forestry, various. a. The shares of agriculture in GDP and in the labor force include forestry and fisheries. 89 90 Distortions to Agricultural Incentives in Asia exceeding US$2,500 in the 1970s, US$5,000 in the 1980s, and US$10,000 in the 1990s.1 The economic growth paths were largely parallel, but, after 1960, Taiwan, China went slightly ahead, with a lead of about five years. It is convenient to review the development of the two economies in four stages, as follows: · Low-income stage (US$1,500 or less), before 1960 · Lower-middle-income stage (US$1,500­US$5,000), 1960­80 · Upper-middle-income stage (US$5,000­US$10,000), 1980­90 · High-income stage (US$10,000 or more), after 1990 The criteria are not applicable to all economies; they are tentatively adopted here for the sake of a comparison in a limited context: the development processes in these two economies.2 Although Korea and Taiwan, China experienced similar changes in real per capita GDP over the four stages, significant differences may be observed in eco- nomic structures. The GDP share of agriculture in Korea in 1955 was nearly 50 percent, whereas, in Taiwan, China, it was below 30 percent. This reflects Korea's greater dependency on agriculture. In both economies, the share of agri- culture in GDP declined significantly (to 3.8 percent in Korea and to 1.7 percent in Taiwan, China by 2004), although Korea's share remained at nearly double the share in Taiwan, China. Historical differences may be observed in the shares of agriculture in the labor force. In 1955, up to 80 percent of workers were employed in agriculture in Korea versus about 50 percent in Taiwan, China. The difference in the share of the labor force employed in agriculture disappeared over time: in 2004, about 7.7 percent of workers were employed in agriculture in the former versus 7.5 percent in the lat- ter. Relatively more rapid declines in the labor force share compared to the GDP share in Korea reflect the urban concentration of industry. In Taiwan, China, there is a wide dispersion of industries over rural areas, and farmers were able to increase their incomes through off-farm employment. While engaging in non- farm activities much of the time, they continued to be classified as farmers. In contrast, in Korea, rural people had to quit farming and migrate to urban areas to obtain nonfarm employment. Such differences are also reflected in the much more rapid decrease in the share of farm household population in the total popu- lation in Korea versus Taiwan, China. The last rows in table 2.1 report the ratios of agricultural GDP per worker to total GDP per worker. This may be considered an indicator of the labor produc- tivity of agriculture relative to the labor productivity of the total economy in nominal value. It may also be regarded as an indicator of the income gap between the agricultural sector and the whole economy. This measure must be interpreted Republic of Korea and Taiwan, China 91 with care. It declined much more quickly in Taiwan, China than in Korea, which may appear to indicate more rapid growth in agricultural labor productivity in Korea. In fact, however, this apparently more rapid relative growth of agricultural labor productivity in Korea was caused by more rapid relative decreases in the size of the farm labor force owing to steadier out-migration of farm labor to urban occupations. The number of agricultural workers decreased more slowly in Taiwan, China because these workers continued to farm, while increasing the allo- cation of their labor to nonfarm activities. Thus, growth in the labor productivity of farmers engaging in agricultural activities relative to that of other workers would not have been slower and might even have been more rapid in Taiwan, China if the ratio is calculated using output per hour of labor instead of output per worker according to the official sectoral labor force classification. Changes in the structures of agriculture How did the structure of agriculture in Korea and Taiwan, China change through the process of economic growth outlined above? Tables 2.1 and 2.2 show that, in 1955, Korea had 2.2 million farm households, which accounted for 62 percent of the total population, while Taiwan, China had 0.73 million farm households, which accounted for 51 percent of the population. The number of farm house- holds has been relatively stable in the latter, whereas it has decreased rapidly in the former. The number of people in farm households in Korea also declined at more rapid rates because of steadier decreases in the number of farm households and the number of persons per farm household. These observations represent addi- tional evidence for the scarcity of nonfarm employment opportunities in rural areas in Korea because of urban-centered industrialization. Indeed, from 1970 to 2004, the share of agricultural income in the total income of farm households declined from 49 to 22 percent in Taiwan, China, whereas, in Korea, it peaked at 76 percent in 1970 and was still 39 percent in 2004. The amount of arable land increased in Korea from about 2 million hectares in 1955 to 2.3 million hectares by 1970, but declined to 1.8 million hectares in 2004. In Taiwan, China, the amount increased from 0.87 million hectares in 1955 to 0.91 million hectares in 1980, but declined to 0.84 million hectares in 2004. Changes in operational farm size in Korea and Taiwan, China during the past five decades have resulted almost exclusively from changes in the number of farm households. In Korea, the average farm size increased from 0.9 hectares of arable land in 1955 to 1.5 hectares in 2004. Meanwhile, the farm size in Taiwan, China remained close to constant at around 1.2 hectares. The much more rapid increase in farm size in Korea was the result of greater out-migration of the farm popula- tion to urban areas owing to more urban-centered industrialization. 92 Table 2.2. Changes in Agricultural Structure, Republic of Korea and Taiwan, China, 1955­2004 Indicator Economy 1955 1960 1970 1980 1990 2000 2004 Farm households, 1,000s Korea, Rep. of 2,218 2,350 2,483 2,155 1,768 1,383 1,240 Taiwan, China 733 786 880 891 860 721 721 Population in farm households, Korea, Rep. of 13,300 14,559 14,422 10,827 6,661 4,031 3,415 1,000s Taiwan, China 4,603 5,373 5,997 5,389 4,289 3,669 3,225 Persons per farm household Korea, Rep. of 6.00 6.20 5.81 5.02 3.77 2.91 2.75 Taiwan, China 6.28 6.84 6.81 6.05 4.99 5.09 4.47 Arable land, 1,000s hectares Korea, Rep. of 1,995 2,025 2,298 2,196 2,109 1,918 1,836 Taiwan, China 873 869 905 907 890 852 836 Arable land per farm household, Korea, Rep. of 0.90 0.86 0.93 1.02 1.19 1.39 1.48 hectares Taiwan, China 1.19 1.11 1.03 1.02 1.03 1.18 1.16 Agricultural income in total farm Korea, Rep. of -- -- 75.8 65.2 56.8 47.2 39.3 household income, % Taiwan, China -- -- 48.7 24.8 20.1 17.6 22.0 Rice in the total value of Korea, Rep. of -- 59.3 37.3 34.1 36.9 32.9 27.6 agricultural production, % Taiwan, China 37.4 36.5 25.7 19.8 12.1 9.6 7.1 Sources: MAF, various; Department of Agriculture and Forestry, various. Note: -- no data are available. Republic of Korea and Taiwan, China 93 This distinguishing characteristic of industrialization in Korea is clearly reflected in the high share of agricultural income in total farm household income. The ratio fell with increases in off-farm employment among members of farm households. In Taiwan, China, the share of agricultural income was already below 50 percent in 1970, when the economy was in the low-income stage, and it fell to about 20 percent in the 1990s, when the economy approached the high-income stage. In contrast, in Korea, the share of agricultural income in total farm house- hold income was 76 percent in 1970 and still nearly 40 percent in 2004; this was not only higher than the share in Taiwan, China, but also higher than the share in Japan at comparable development stages. Major differences in the adjustments undertaken in agriculture in the face of economic growth based on industrial development are also observable in the changes in the commodity mix in farm production. Rice was traditionally the most important crop in both economies, and this importance tended to decline as per capita income rose. However, the changes in the relative importance of rice was different in the two economies. In Korea from 1960 to 2004, the share of rice in the total value of agricultural production declined from 59 to 28 percent. In Taiwan, China, the share was initially low, at 37 percent in 1960, but fell more rap- idly, to only 7 percent in 2004. The difference reflects the greater opportunity to grow tropical cash crops there. It also reflects the economy's success in achieving greater agricultural diversification toward high-valued commodities such as veg- etables, fruits, chicken, and pork in response to the shift in demand toward more income-elastic commodities. The Evolution of Agricultural Policy We now outline changes in agricultural policies over the different stages of devel- opment in the two economies. Korea Before 1960, Korea was a low-income country, with a per capita income below US$1,500.3 The economy had been severely damaged during the Korean War. The agricultural policy adopted at this stage aimed to maintain low domestic con- sumer prices for staple foods, notably rice and barley, as well as for fertilizer. The Grain Management Law, enacted in 1950, gave the government the authority to regulate price on staple foods. However, government control was not effective during the 1950s, given that the market share of government-controlled rice was less than 10 percent. The government was supposed to purchase grain directly from farmers, but it was unable to purchase sufficient amounts; there were 94 Distortions to Agricultural Incentives in Asia budgetary constraints, and inflation had caused grain prices to spiral upward in the mid-1950s. Schemes to collect rice in payment of the land tax and to barter fertilizer for rice were initiated. The first type of scheme was successful, but the second type failed because the implicit barter price of rice was lower than the market price. Grain imports from the United States under Public Law 480, which amounted to 8 to 12 percent of total domestic grain in 1956­65, helped the gov- ernment keep grain prices low. In the 1960s, under the development autocracy of Pak Chong-hui, Korea launched an energetic policy to promote industrialization. Rather than support- ing adequate incomes among farmers, agricultural policies at the time were designed to keep prices on staple food crops low so as to maintain low wage rates and a low cost of living among industrial workers. Government purchase prices were set below market prices; this was considered necessary to raise industrial profits and foster capital formation. The government's price interventions gradu- ally became more intense. The market share of government-controlled rice was expanded to 20­25 percent during the 1960s; this was used principally to main- tain low domestic prices. These agricultural-taxing policies continued during the initial part of the lower-middle-income stage. The Korean economy advanced quickly toward the upper-middle-income stage. Meanwhile, the direction of agricultural policy gradually moved toward the provi- sion of support for farmers. In the early 1970s, the buffer-stock operation for non- cereal products was set in motion to counteract price declines in the subsector. Chemical fertilizers, pesticides, and farm machinery were added to the list of subsi- dized inputs (alleviating the adverse impact on farmers of the import protection provided to the manufacturers of these inputs). The government's purchase prices for rice and barley were steadily raised with the aim of increasing food production, as well as reducing the urban-rural income gap. Although the government allowed an increase in the producer prices on staple food grains, this occurred without a comparable rise in the market prices for rice and barley, thereby preventing a rise in the cost of living and the wage rates among industrial workers. Likewise, the gov- ernment assisted livestock producers in part by using import quotas rather than tariffs to protect them from import competition; the rents from the quotas were captured by the producer-managed meat import agency.4 The implementation of the two-price system, however, conflicted with the need to maintain financial and monetary stability. When the difference between the pur- chase prices and the sale prices of rice and barley widened, the deficit in the grain management fund increased. Because a large portion of the deficit was financed through long-term overdrafts at the Bank of Korea, the policy came to represent a major addition in inflationary pressure. The expansion in the government deficit caused by the two-price policy became a serious constraint on the policy. Republic of Korea and Taiwan, China 95 When the economy entered the upper-middle-income stage in the 1980s, the government took a step toward reducing both tariff and nontariff protection for manufacturing industries. In contrast, agricultural policies that tended to protect farmers were strengthened. The producer prices of farm products were raised far above border prices by means of quantitative import restrictions on most agricul- tural commodities. After Korea entered the high-income stage in the early 1990s, significant policy changes were mostly related to the Uruguay Round Agreement on Agriculture, which was stipulated in 1995. According to the provisions of the agreement, Korea's quantitative restrictions were converted to tariffs for all agricultural prod- ucts except rice. In the Uruguay Round negotiations, Korea retained the status of a developing country, which assigned it special treatment in the implementation of the commitments to reduce border protection. The agricultural products under tariffication were subject to a protection reduction commitment of 24 percent, on average, within 10 years, and a minimum cut of 10 percent. The tariff rates on Korean agricultural products were over 60 percent, on average. The tariffs on products that were considered particularly important in Korea were cut by the minimum rate of 10 percent. In addition, many agricultural products began to be imported under the minimum market access commitment. This commitment required that, for all agricultural products, at least 3 percent of consumption must be purchased inter- nationally during the first year, and the import share must increase annually up to 5 percent of consumption within 10 years. Low tariff rates were applied to the in- quota volume so as to guarantee easy market access from exporting countries. Many key agricultural products such as rice, barley, oranges, red peppers, garlic, and onions were newly imported under this commitment. Rice, the most important item in Korean agriculture, was temporarily exempted from tariffication as provided in the Uruguay Round Agreement on Agriculture, annex 5.B. As an exception, rice was subject to an import quota, beginning at 1 percent of total consumption and gradually increasing up to 4 per- cent in 2004, the final implementation year. If Korean rice had not been exempted from tariffication, Korea would have complied with the standard market access commitment, 3 to 5 percent. The temporary exemption from tariffication expired in 2004, but Korea opted to continue invoking a rice exemption from tariffication for another 10 years, to 2014. Taiwan, China After World War II, Taiwan, China suffered high inflation rates, serious shortages of food and other necessities, and a heavy defense burden.5 The government gave 96 Distortions to Agricultural Incentives in Asia the highest priority to economic stabilization, food production increases, and the repair of war damage. To alleviate the intense population pressure on the limited land, it decided to grant incentives to farmers. Together with the land reform pro- gram implemented between 1949 and 1953, war-damaged irrigation and drainage facilities were repaired, fertilizers and other farm inputs were made available, and farmer organizations were strengthened. During the recovery stage of the economy, the Sino-American Joint Commis- sion on Rural Reconstruction, established in Nanjing in 1948, played an impor- tant role. The commission served as a nonpermanent agency for the postwar rural reconstruction of China. From 1951 to 1965, the United States provided US$1.5 bil- lion in aid in Taiwan, China through the commission. Approximately one-third of the aid went to agriculture, where the aid was used to build infrastructure and enhance human resources. Also, substantial imports of commodities financed by U.S. aid and increases in domestic production, especially food production, helped relieve demand pressure. During the low-income stage of economic development (before 1960), agricul- tural policy in Taiwan, China was designed mainly to supply rice to the nonfarm population at low, stable prices. In those days, two important taxes were imposed on farmers: the farm land tax and the hidden rice tax. Implementation was achieved by means of compulsory rice purchases and a rice-fertilizer barter system. The compulsory purchase of paddy from landowners at official prices was another method used to exert government control over rice. All paddy lands were subject to the paddy land tax, plus the compulsory procurement of rice. The compulsory pro- curement was assessed on the basis of tax units determined according to land pro- ductivity. The difference between the government procurement prices and farmer market prices constituted a hidden tax on paddy landowners, who, after the land reform program, were mostly farm operators. The hidden tax was gradually reduced as per capita incomes rose, but it continued to be imposed until 1973. The government's rice collection by all these means during 1950­70 averaged 50 to 60 percent of the total amount of rice produced, minus the home consumption of farm families. By 1973, however, the share had declined to 20 percent. In subse- quent years, it rose again because of the launch of the guaranteed rice price policy. The total value of the hidden rice tax was more than twice the value of the farm land tax before 1961 (except in 1954) and was larger than the total income tax before 1963. After 1961, when the economy moved to the lower-middle-income stage, the hidden rice tax decreased rapidly: in 1971, the ratio of the hidden rice tax to the total income tax was only 8.5 percent (Kuo 1975). Agricultural policy geared to exploit agriculture for the sake of supporting industrial development (and military development) largely ended during the 1970s, when the shift toward the subsidization of agriculture was initiated. This Republic of Korea and Taiwan, China 97 was the period when labor-intensive light industries were rapidly expanding in response to increases in export demand. Because many light industries such as garments and footwear were located in rural areas, nonfarm incomes became increasingly important among farm households. Farmers were able to take advan- tage of employment in manufacturing without leaving home; and, moreover, many farmers also engaged in nonfarm self-employed activities during less busy farm seasons. Therefore, the need for farmers to rely on agricultural protection policies was less substantial in Taiwan, China than in Korea. In 1978, Taiwan entered the upper-middle-income stage at a real GDP per capita above US$5,000. To help raise the incomes of farm workers to levels comparable to the incomes of workers in the rapidly expanding industrial sector, the government offered loans and subsidies to promote farm mechanization; the scheme was designed to raise farm labor productivity. At this time, the growth in rice production began to slow in response to renewed emphasis on livestock and fishery products and high-value export crops. Increases in industrial employment were also pushing up the costs of farm labor. Agriculture continued to lag behind the industrial sector in labor productivity, and the gap between farm and nonfarm per capita incomes was widening, to the disadvantage especially of farmers who relied mainly on rice production. The problems faced in Taiwan, China in agriculture were similar to those experienced by many industrial countries at comparable stages of develop- ment, especially Japan in the early 1960s and Korea in the late 1970s. The annual per capita consumption of rice in Taiwan, China fell from 140 to 74 kilograms between 1968 and 1988. The accumulation in rice reserves became a serious problem. To reduce production, farm extension workers encouraged farmers to plant other crops in rice fields. The effort was unsuccessful because no economic incentive was provided. A six-year rice-crop substitution plan was inau- gurated in 1984 that involved direct subsidies of 1 metric ton of paddy rice per hectare to farmers who shifted their rice fields to corn or sorghum, or 1.5 metric tons of paddy rice per hectare to farmers who shifted to crops other than corn and sorghum. In addition, corn and sorghum were purchased by the government at guaranteed prices. Under the program, rice production declined to 1.8 million metric tons in 1988, which was 0.9 million metric tons less than the peak, which had been reached in 1976. The in-kind subsidy was altered to a cash payment in 1988 to improve efficiency in the management of the program. The economy entered the high-income stage in the late 1980s. Real GDP per capita exceeded US$10,000 beginning in 1988. The most important changes in agricultural policy during the high-income stage were related to the accession of Taiwan, China to the World Trade Organization, which became effective on January 1, 2002. In line with the level of development of the economy, Taiwan, China agreed to adjust its tariff rates so that they were between the rates in Japan 98 Distortions to Agricultural Incentives in Asia and the rates in Korea. It also agreed to reduce tariffs from the average nominal rate of 20 percent in 2001 to 14 percent in the first year of accession and to 13 per- cent by 2004. The target date for tariff reductions was 2002 except for 137 items that were under tariff rate quotas. Of the 41 products that were under import quota restrictions before accession to the World Trade Organization, 18 were moved to tariffication after accession. Rice received a special exemption, and the remaining 22 items were governed by the tariff rate quota regime. As in the case of Korea, the special treatment of rice is based on the rules of the Uruguay Round Agreement on Agriculture, annex 5. The quota of rice imports was set in 2002 at 8 percent of the average domestic consumption between 1990 and 1992 (144,720 tons of brown rice). By negotiation, this amount was divided into governmental and private import quotas. The government rice quota (65 percent of rice imports) was subject to the same treatment as rice purchased from local growers. The imported rice cannot be exported for food aid, nor can it be used for animal feed. The remainder (35 percent) was imported by private firms and was allocated on a first-come-first-served basis. For both private and government quotas, there is a ceiling on the price markup of NT (New Taiwan) $23.26 per kilogram for rice and NT$25.59 for rice products sold on the domestic market. If the sale of quota rice is slow, the price markup can be cut by NT$3 every two weeks. The markup reduction may be continued until all quantities are sold. The Measurement of Distortions to Agricultural Incentives The main focus of this study is the measurement of the extent of distortions to agricultural prices brought about by government-imposed policy measures. The measures examined are those that create a gap between domestic prices and prices as they would be under free markets. Since it is not possible to understand the characteristics of agricultural development from a sectoral perspective alone, the project's methodology estimates the effects of direct agricultural policy measures (including any distortions in the foreign exchange market) and, for comparative evaluation, also generates estimates of distortions in nonagricultural sectors. Specifically, our study computes NRAs for farmers that reflect an adjustment for direct interventions on inputs, such as border protection for fertilizers. Via the calculation of an RRA, it also generates NRAs for nonagricultural tradables for comparison with the NRAs for agricultural tradables (see appendix A). The commodities for which we calculate NRAs include rice, wheat, barley, beef, soybeans, pig meat, poultry, eggs, and milk for Korea. For Taiwan, China we esti- mate NRAs for rice, wheat, beef, pig meat, poultry, and eggs. Domestic prices have been converted to U.S. dollars using market rates for foreign exchange rates except Republic of Korea and Taiwan, China 99 for 1955­64 in Korea and 1955­61 in Taiwan, China. The shadow exchange rates estimated for Korea by Frank, Kim, and Westphal (1975) and for Taiwan, China by Scott (1979) are used to take into account the distortions in foreign exchange markets during those earlier years. Aggregate NRAs for output for each county are calculated using weights based on the domestic production of commodities val- ued at undistorted prices. In addition to the commodities covered in this study, three other crops-- peppers, garlic, and Chinese cabbages--are included in the NRA calculations for Korea. The estimates for these products come from the producer and consumer support estimates in the OECD PSE-CSE Database (2007). The data for these crops are available only beginning in 1986. We assume that the distortions affect- ing these crops prior to 1986 were equivalent to about 90 percent of the distor- tions affecting the available products that are covered. Valued at distorted prices, the share of agricultural output surveyed in this study is between 50 and 70 percent for Korea and somewhat less for Taiwan, China. It is difficult to judge the NRAs for the residual products that are not cov- ered. We assume they are made up of the following share trends (at distorted prices) between 1955 and the present: import-competing at 50 to 80 percent and nontradables at 50 to 20 percent in Korea. The distortions to the residual products are assumed at zero for nontradables and, to import-competing products, at the same level applied in the four products in the Organisation for Economic Co-operation and Development study. For Taiwan, China, we assume that the dis- tortions of all the noncovered residual products are zero because most of these products are nontradable or exportable. To compute the RRAs, we estimate the NRAs for nonagricultural industries. For the latter, weighted tariffs were available for the two economies only for selected years. We linearly interpolated data for the years for which data are not available. For the early years, the tariff rates are estimated as the value of total tar- iff revenue, divided by the value of imports. Assuming the exportable industries receive no assistance, we then multiply the weighted average tariff by the share of import-competing industries in the value of all nonagricultural tradables. This procedure undoubtedly underestimates the assistance to nonagricultural indus- tries (and thereby the implicit taxation of agriculture) in the 1950s and, to a lesser extent, the 1960s. This is so because it does not account for the nontariff import restrictions that were rife, nor for Korea's subsidized credit for target industries. The estimation results for the NRAs on the covered farm products are summa- rized as five-year averages in tables 2.3 and 2.4 for Korea and Taiwan, China, respectively, while table 2.5 reports the RRAs for both economies.6 The annual movements in the NRAs and RRAs are shown separately for the two economies in figures 2.1 to 2.6. 100 Table 2.3. NRAs for Covered Agricultural Products, Republic of Korea, 1955­2004 (percent) Product, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Exportablesa n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Import-competing productsa 3.9 4.4 16.6 47.6 73.8 122.8 166.7 201.9 182.9 213.6 Wheat 43.0 26.7 11.2 0.4 26.5 92.2 144.4 216.0 122.8 135.4 Barley 41.2 83.5 72.3 120.3 101.2 165.9 357.0 524.3 543.0 562.8 Rice 8.2 7.0 5.4 31.3 59.6 118.4 214.4 265.9 294.3 385.9 Beef 38.8 34.4 64.9 73.9 162.6 163.2 126.2 200.8 159.9 167.8 Pig meat 15.2 21.7 158.7 204.1 202.9 169.1 124.7 149.3 116.2 134.4 Poultry 11.8 6.9 131.4 103.5 161.7 94.2 86.6 155.6 171.7 179.2 Eggs 32.3 24.7 23.0 0.1 7.5 14.9 19.4 28.0 26.6 54.3 Milk -- -- 173.3 108.8 189.0 179.8 185.2 203.7 140.7 149.8 Cabbages -- -- -- -- -- -- 30.0 30.0 29.1 27.6 Peppers -- -- -- -- -- -- 175.0 245.4 145.5 197.0 Soybeans 13.0 18.8 58.8 80.0 122.2 253.0 361.8 508.2 625.6 757.4 Garlic -- -- -- -- -- -- 250.3 288.8 213.3 122.6 Total, covered productsa 3.9 4.4 16.6 47.6 73.8 122.8 166.7 201.9 182.9 213.6 Dispersion, covered productsb 33.4 40.0 82.7 81.0 87.2 64.7 112.4 164.2 200.1 225.4 % coverage, at undistorted prices 45 55 64 61 61 56 60 57 52 46 Source: Honma and Hayami 2007a. Note: -- no data are available; n.a. not applicable. a. Weighted averages; weights are based on the unassisted value of production. b. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of the covered products. Table 2.4. NRAs for Covered Agricultural Products, Taiwan, China, 1955­2002 (percent) Product, indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­02 Exportablesa 23.5 7.5 5.7 20.7 13.4 35.9 89.5 161.4 167.6 203.1 Rice 29.6 6.6 17.9 9.4 7.6 32.5 103.3 161.4 167.6 203.1 Pig meatb 8.1 64.0 99.7 98.3 60.6 42.6 64.8 n.a. n.a. n.a. Import-competing productsa 33.0 5.3 21.7 26.7 32.5 49.1 55.4 93.6 126.3 160.0 Wheat 48.2 36.0 39.4 32.2 57.2 92.3 -- -- -- -- Beef 13.7 41.2 28.8 22.0 79.6 77.0 101.3 98.5 82.6 72.8 Pig meatb n.a. n.a. n.a. n.a. n.a. n.a. n.a. 107.1 131.3 173.2 Poultry 47.5 3.7 21.2 27.1 30.0 63.6 84.6 143.0 228.7 279.5 Eggsc n.a. n.a. n.a. n.a. n.a. 0.7 26.8 23.9 17.9 24.7 Nontradablesa 0.0 0.0 0.0 0.0 0.0 0.0 n.a. n.a. n.a. n.a. Eggsc 0.0 0.0 0.0 0.0 0.0 0.0 n.a. n.a. n.a. n.a. Total, covered productsa 23.2 7.2 6.2 20.0 14.0 35.1 76.1 109.5 134.0 167.8 Dispersion, covered productsd 33.4 35.3 47.5 40.5 40.5 34.5 56.9 66.1 86.9 106.4 Coverage, at undistorted prices, % 53 49 49 48 50 42 35 34 35 36 Source: Honma and Hayami 2007a. Note: -- no data are available; n.a. not applicable; the applicable data are shown elsewhere in the table under another trade status. a. Weighted averages; weights are based on the unassisted value of production. b. In 1989, the trade status of pig meat was changed from import-competing to exportable. The period average reported here corresponds to 1985­88 for the import-competing product and to 1989­94 for the exportable product. c. Eggs were assumed to be a nontradable, with zero distortions, prior to 1983. d. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of the covered products. 101 Table 2.5. NRAs in Agriculture Relative to Nonagricultural Industries, Republic of Korea and Taiwan, China, 1955­2004 102 (percent) Indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Korea, Rep. of Covered productsa 3.9 4.4 16.6 47.6 73.8 122.8 166.7 201.9 182.9 213.6 Noncovered products 1.7 0.2 7.0 15.3 25.3 37.4 64.3 88.0 74.6 71.7 All agricultural productsa 3.2 4.0 13.4 35.7 56.3 89.4 126.1 152.8 129.8 137.3 Non-product-specific assistance -- -- -- -- -- -- -- -- -- -- Total agricultural NRAb 3.2 4.0 13.4 35.7 56.3 89.4 126.1 152.8 129.8 137.3 Trade bias indexc n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. NRA, all agricultural tradables 3.3 4.9 16.3 46.1 71.8 118.6 159.3 197.6 164.8 171.9 NRA, all nonagricultural tradables 45.6 37.1 22.3 11.4 11.7 6.8 5.7 3.3 2.3 1.7 RRAd 32.6 21.4 4.8 30.5 53.9 104.8 145.5 188.2 158.8 167.3 Indicator 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­02 Taiwan, China Covered productsa 23.2 7.2 6.2 20.0 14.0 35.1 76.1 109.5 134.0 167.8 Noncovered products 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 All agricultural productsa 11.8 3.5 3.0 9.2 7.0 14.6 26.4 37.2 45.5 60.0 Non-product-specific assistance -- -- -- -- -- -- -- -- -- -- Total agricultural NRAb 11.8 3.5 3.0 9.2 7.0 14.6 26.4 37.2 45.5 60.0 Trade bias indexc 0.15 0.05 0.02 0.12 0.05 0.15 0.27 0.11 0.02 0.00 NRA, all agricultural tradables 15.8 4.7 3.9 12.0 8.9 18.5 32.7 45.0 53.6 69.2 NRA, all nonagricultural tradables 8.8 9.3 8.8 7.5 7.0 5.2 4.5 2.6 1.8 1.1 RRAd 22.5 4.2 4.5 4.2 1.7 12.7 27.0 41.3 51.0 67.3 Source: Honma and Hayami 2007a. Note: -- no data are available; n.a. not applicable. a. Including product-specific input subsidies. b. Including product-specific input subsidies and non-product-specific assistance. The total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. c. The trade bias index is defined as (1 NRAagx/100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. d. The RRA is defined as 100*[(100 NRAagt)/(100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. Republic of Korea and Taiwan, China 103 The estimated RRAs for the two economies in the 1950s and 1960s--during the low-income and lower-middle-income stages of development--were low; the RRAs involved negative rates for most years before the 1970s. If one were able to include in the calculations the nontariff barriers limiting industrial imports dur- ing the period, the resulting RRAs would have been even more negative. In any case, the RRAs suggest that, in Korea in the later 1950s, the relative prices for out- puts faced by farmers were over 30 percent below the output prices they would have faced under free-market conditions, while they were over 20 percent below in Taiwan, China. Output prices were still low in the 1960s, when RRAs averaged about 15 percent in Korea and 5 percent in Taiwan, China (table 2.5 and fig- ure 2.1). In Korea, there was a rapid increase in agricultural assistance and a decline in the protection for manufacturing beginning in the 1970s when the economy moved from the lower- to the upper-middle-income stages.7 In Taiwan, China, the transition was similar, but less rapid. Thus, the RRA lines in figure 2.1 crossed in the late 1960s, and--apart from 1974, when world food prices spiked-- the RRAs for Taiwan, China thereafter remained below and grew less rapidly than the RRAs for Korea, although Korea lagged slightly in terms of per capita income. Figure 2.1. RRAs for Agricultural and Nonagricultural Tradables, Japan, Republic of Korea, and Taiwan, China, 1955­2004 250 200 150 % 100 RRA, 50 0 50 100 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 year Korea, Rep. of Japan Taiwan, China Source: Honma and Hayami 2007b. Note: For the definition of the RRA, see table 2.5, note d. 104 Distortions to Agricultural Incentives in Asia Figure 2.2. NRAs for Agricultural Products, Republic of Korea and Taiwan, China, 1955­2004 200 150 100 % NRA, 50 0 50 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 year Korea, Rep. of Taiwan, China Source: Honma and Hayami 2007a. Even after the two economies entered the high-income stage of development in the 1990s, the trend continued in the rising RRAs, reaching nearly 170 percent for Korea (above the 125 percent for Japan) and almost 70 percent for Taiwan, China in the first years after 2000 (table 2.5). Figure 2.2 shows the movements in the weighted average NRAs for agricultural products in the two economies. They are similar to the movements in the RRAs in both economies after the 1970s because the impact of nonagricultural NRAs on the RRA calculation became trivial. For Korea, the NRAs refer only to import-competing products because there have been no significant farm exports from Korea for decades. Two additional issues are worthy of note in the case of Taiwan, China. First, tradable products show estimated NRAs that are higher than the NRAs of the sector as a whole. This is because we assume that the products not covered explicitly in our study are exportable or nontradable and show a zero average NRA. Second, in Taiwan, China, the exportables covered (predominantly rice) show average NRAs that are higher than the average NRAs for the import-competing products covered (fig- ure 2.3). This generates the positive trade bias index shown in table 2.5 and is possibly a unique situation; typically, governments find it much easier politically to provide assistance to import-competing industries (via import restrictions) rather than to exporting industries. Republic of Korea and Taiwan, China 105 Figure 2.3. NRAs for Exportable, Import-Competing, and All Agricultural Products, Taiwan, China, 1955­2002 250 200 150 % 100 NRA, 50 0 50 100 1955 1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 year import-competing total exportables Source: Honma and Hayami 2007a. There were wide fluctuations in the RRAs for Korea in the late 1990s because of the currency crises in Asia that began in 1997. The sharp increases in the RRAs for Taiwan, China in 1999 and 2000 were caused by shortages in livestock prod- ucts after the September 1999 earthquake and the reductions in the production of pig meat resulting from the spread of foot-and-mouth disease among pigs in 1997. Fluctuations in the RRAs and NRAs arose mainly from changes in the NRAs of individual farm commodities and changes in the weights of each commodity (fig- ure 2.4). In both economies, the most important agricultural product was rice. The protection of rice therefore exercised a large influence on the RRAs. The movements in the NRAs for rice in the two economies, shown in figure 2.5, are clear: a strong upward trend from the mid-1970s as both economies began to move out of the lower-middle-income stage. Over the past three decades, the weight of rice in agricultural production in Korea has been twice the correspon- ding weight in Taiwan, China (see table 2.2). Korea was exempted from the tariffi- cation of rice under the Uruguay Round Agreement on Agriculture; this allowed rice NRAs to grow despite Korea's commitment to the World Trade Organization to lower the protection on farm products. 106 Distortions to Agricultural Incentives in Asia Figure 2.4. NRAs for Agricultural and Nonagricultural Tradables and the RRA, Republic of Korea and Taiwan, China, 1955­2004 a. Republic of Korea 250 200 150 100 percent 50 0 50 100 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 year NRA, agricultural tradables NRA, nonagricultural tradables RRA b. Taiwan, China 100 80 60 40 cent 20 per 0 20 40 60 1955 1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 year NRA, agricultural tradables RRA NRA, nonagricultural tradables Source: Honma and Hayami 2007a. Note: For the definition of the RRA, see table 2.5, note d. Republic of Korea and Taiwan, China 107 Figure 2.5. NRAs for Rice, Republic of Korea and Taiwan, China, 1955­2004 500 400 300 % 200 NRA, 100 0 100 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 year Korea, Rep. of Taiwan, China Source: Honma and Hayami 2007a. In the two economies, support for farmers has been mostly provided through restrictions on food imports. In addition, there have been schemes whereby crop producer prices have been supported above the prices charged to grain and soy- bean consumers, including feedmixers that supply livestock producers with ani- mal feedstuffs. Thus, the consumer tax equivalents for food are below the NRAs for some crop products. This, combined with the different weights of various products in consumption and production, has meant that, for Korea in 2000­04, the average NRA for the products covered is more than 30 percent above the con- sumer tax equivalent, while, for Taiwan, China in 2000­02, it is more than 65 per- cent above the consumer tax equivalent (compare tables 2.3, 2.4, and 2.6). Thus, consumers have been spared some of the implicit tax that otherwise would have been imposed on them had border measures alone been used to raise producer prices above international levels. Sources of Growth in Agricultural Protection The experiences of Korea and Taiwan, China offer good illustrations of the policy shift from the exploitation of agriculture to the protection of agriculture as economies grow and industrialize. In these two economies, agricultural protection 108 Table 2.6. CTEs for Covered Agricultural Products, Republic of Korea and Taiwan, China, 1955­2004 (percent) a. Republic of Korea Product 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Wheat 46.2 22.5 11.1 1.1 16.2 46.0 132.8 167.4 80.5 80.6 Barley 40.8 77.8 64.9 96.6 57.3 119.6 325.6 411.5 341.2 327.5 Rice 7.7 5.5 5.0 29.1 54.5 113.4 211.5 261.7 290.8 385.3 Beef 38.8 34.4 64.9 73.9 162.6 163.2 122.1 200.7 153.9 167.7 Pig meat 15.0 21.7 158.7 204.1 202.9 169.1 124.7 149.3 116.2 134.4 Poultry 11.6 7.0 131.4 103.5 161.7 94.2 86.6 155.6 171.7 179.2 Eggs 27.1 24.7 23.0 0.1 7.5 14.9 19.4 28.0 26.6 54.3 Milk -- -- 130.0 108.8 189.0 179.8 185.2 203.7 140.7 149.8 Soybeans 19.8 8.2 51.6 63.2 95.2 245.4 112.2 75.5 63.6 66.8 Total, covered productsa -- 6.6 13.6 39.2 63.3 110.6 150.8 185.9 159.6 163.9 b. Taiwan, China Product 1955­59 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­02 Rice 29.6 6.6 17.9 9.4 7.6 32.5 103.3 161.4 167.6 203.1 Wheat 33.9 23.5 29.3 14.7 1.1 40.3 183.4 -- -- -- Beef 13.7 41.2 28.8 22.0 79.6 77.0 101.3 98.5 82.6 72.8 Pig meat 8.1 64.0 99.7 98.3 60.6 42.6 64.8 107.1 131.3 173.2 Poultry 47.5 3.7 21.2 27.1 30.0 63.6 84.6 143.0 228.7 279.5 Eggs 0.0 0.0 0.0 0.0 0.0 0.7 26.8 23.9 17.9 24.7 Total, covered productsa 26.2 27.0 21.7 23.2 18.6 39.9 82.6 116.4 136.7 99.8 Source: Honma and Hayami 2007a. Note: CTE consumer tax equivalent. -- no data are available. a. Weighted averages; weights are based on the undistorted value of consumption. For the Republic of Korea, this includes cabbages, peppers, and garlic. Republic of Korea and Taiwan, China 109 levels were negative in the 1950s and the 1960s and began to rise sharply in the late 1970s following successes in industrial development. Anderson, Hayami, and Honma (1986) found the growth of agricultural pro- tection during the three decades to the early 1980s far more rapid in these economies than in the pioneers of industrialization in the West or even Japan. Moreover, the two economies were not exceptional in their bent toward growth in agricultural protection, and the growth was unusually rapid because the industrial development and the associated structural changes were also unusually rapid. The growth in protection continued at the same pace to the early 1990s, when the two economies were in the middle-income stage, but it began to decelerate thereafter, particularly in Korea when that economy reached the high-income stage. The rapid growth in agricultural protection is largely explained by the shift in comparative advantage away from agriculture to industry as the result of success- ful industrialization. This shift was most pronounced during the middle-income stage. Relative to population, the two economies are characterized by meager endowments of natural resources, including land for cultivation. Industrialization was the only possible route to sufficient economic growth to catch up with advanced economies. Economic growth through the exploitation of abundant natural resources--vent-for-surplus growth (Myint 1965)--was not an option in these economies, unlike in Southeast Asian countries such as Malaysia and Thai- land (Douangngeune, Hayami, and Godo 2005). The decline in agriculture's com- parative advantage because of successful industrialization increased intersectoral resource adjustment costs and led to a widening rural-urban income disparity that, if left to the solutions of competitive markets, would have been shouldered by farmers. This raised demands among farmers for agricultural protection. The association between the rise in agricultural protection and the decline in agriculture's comparative advantage was tested by Honma and Hayami (1986) using multiple regression analysis and a pooled data set on 15 countries at six points of time ending in 1980. They found a strong correlation between the aggre- gate nominal rate of protection and the index of agriculture's labor productivity relative to the total economy's labor productivity.8 Based on the results, Honma and Hayami concluded that the high level of agricultural protection in East Asia was caused not so much by factors unique to East Asia, but mainly by factors com- mon to all industrial countries. One may argue that high rates of agricultural protection hinder adjustments aimed at raising agricultural productivity and thereby slow the growth in agri- culture's comparative disadvantage. However, it is doubtful that the needed adjustments in agriculture in the face of the rapid industrial development experi- enced by East Asia's top performers could have been achieved sufficiently quickly under free markets to avoid major disruptions and social instability. This is 110 Distortions to Agricultural Incentives in Asia particularly so because, in these economies, the agrarian structure involves a large number of small, independent family farms possessing unfavorable resource endowments. Korea and Taiwan, China were at the low-income stage of economic develop- ment in the 1950s and entered the middle-income stage in the 1960s. During the middle-income stage, productivity growth in agriculture lagged relative to nona- griculture because of successful industrialization. Farm incomes declined relative to the incomes of nonfarm households. Nonetheless, during the lower-middle- income stage, it was not possible for the government to secure sufficient financing to close the income gap because the share of agriculture in the national income and in the labor force was still large. Thus, before Korea and Taiwan, China reached the upper-middle-income stage, agricultural protection remained low despite growing rural-urban income disparity. The basic agricultural problem confronted by middle-income economies such as these in the 1960s and 1970s is known as the disparity problem (Hayami and Godo 2004; Hayami 2005). In this case, it revolves around the income disparity between farm and nonfarm households brought about by a lag in productivity growth in agriculture relative to nonagriculture. This lag was caused by the indus- trialization that successfully raised these economies to the middle-income stage. With respect to the low-income stage, food supply capacity rises during this stage because of both the productivity growth in domestic agriculture and the higher foreign exchange earnings, which enable more food imports. Meanwhile, factors causing growth in the demand for food, such as high population growth and high demand elasticities, are weakened. The terms of trade between agriculture and nonagriculture thus worsen. Farmer incomes tend to decline relative to nonfarmer incomes. This reflects the widening intersectoral productivity gap. Observing the rapid escape of non- farm workers from poverty, farmers begin to realize the relative extent of their poverty even if their incomes have not fallen or have risen only slightly since the previous stage. The dissatisfaction of these farmers who have remained poor often becomes a significant source of social instability. Thus, at the middle-income stage, preventing growth in rural-urban income disparity is a prime concern of policy makers. To achieve this goal, the government might adopt agricultural protection measures to appease farmers so that their dis- satisfaction is not transformed into a serious antigovernmental movement. This protection may not be sufficient, however, to close the income gap between farmers and urban workers until the country graduates from the lower middle-income stage. Because the share of agriculture in national income and in the labor force is still large, it is difficult to raise sufficient revenue from the nonfarm sectors to close the growing farm-nonfarm income gap through direct Republic of Korea and Taiwan, China 111 support payments. It is also difficult to pass on the cost of agricultural protection to consumers by raising food import barriers; increases in food prices erode the real wages paid by the large number of small-scale enterprises that rely heavily on cheap labor. Faced with the disparity problem, policy makers in middle-income countries are forced to search for ways to protect farmers within the constraints of the food problem. The food problem binds because a large number of urban workers have remained absolutely poor and food still accounts for a high share in their house- hold expenditures. Per capita incomes rose rapidly in the late 1970s and early 1980s as Korea and Taiwan, China approached the upper-middle-income stage. Both the govern- ment and the nonagricultural sectors became capable of shouldering the cost of agricultural protection during the upper-middle-income stage. It may be argued that high protection contributed to the loss in agricultural comparative advan- tage in these two economies by reducing the incentives to improve agricultural productivity. The effect of high protection on comparative advantage was much smaller than the effect of industrial productivity growth, which was unusually rapid among the high performers in East Asia. The conditions necessary for growth in agricultural protection were created by the increases in industrial productivity growth, which occurred far more quickly than agricultural productivity growth and provided the higher earnings capable of supporting agriculture. There were also differences in the policy approaches toward agricultural pro- tection in the two economies. During the upper-middle-income stage, agricul- tural protection rose more quickly and to higher levels in Korea. This seems to reflect the differences in the cost of the intersectoral adjustments required by farmers because of the shifts in comparative advantage. In Korea, the shift of labor from agriculture to nonagriculture involved the migration of workers from rural to urban areas; whereas, in Taiwan, China, much of the shift was absorbed by farm households, which increased their nonfarm activities in their home villages and towns. The pecuniary and psychological costs of the intersec- toral labor reallocation would have been correspondingly higher in the farm sector in Korea. The ratio of agricultural GDP per worker to total GDP per worker in nominal terms decreased more rapidly in Taiwan, China (see table 2.1). This does not mean that agricultural adjustment occurred more slowly there. One should make allowance for the fact that output per hour of labor in agricultural activities differs from output per farm worker as carried in official statistics. Since farmers in Taiwan, China depended more heavily on earnings from nonfarm activities (table 2.2), it is likely that household incomes per capita were not lower, but 112 Distortions to Agricultural Incentives in Asia Figure 2.6. RRAs in Agriculture and Relative GDP per Agricultural Worker, Republic of Korea and Taiwan, China, 1955­2004 200 150 100 % RRA, 50 0 50 10 20 30 40 50 60 70 ratio of agricultural GDP per worker to total GDP per worker, % Korea, Rep. of Taiwan, China Source: Honma and Hayami 2007a. Note: Calculated at 2000 constant prices in U.S. dollars. higher than the corresponding incomes in Korea, although, in Taiwan, China, GDP per farm worker was lower relative to GDP per worker in the total economy. The negative relationship between the relative GDP per worker and the RRA in the two economies in 1955­2004 is shown in figure 2.6.9 Korea intensified agricul- tural protection despite the relatively high incomes among farmers. In addition to the high cost of intersectoral adjustment shouldered by Korean farmers, the con- stant menace of Communist aggression from the north caused citizens to feel the need to support farmers to maintain political stability. In their analysis, Honma and Hayami (1986) found that political power is maximized in the agricultural sector when the share of agriculture declines to 4 to 5 percent of GDP or to 5 to 8 percent of the labor force. Korea recently entered this peak zone in terms of both GDP and labor force. Taiwan, China recently entered it in terms of the labor share (having entered this zone in 1990 in terms of GDP share). Political economy factors may underlie the rise in agricultural protection- ism in Korea at the high-income stage since 1990, as observed through NRAs at the farmgate, despite the lack of an apparent additional increase in agricultural comparative disadvantage.10 Republic of Korea and Taiwan, China 113 Concluding Remarks This chapter examines changes in the distortions to agricultural incentives in terms of price distortions in Korea and Taiwan, China in a manner consistent with the methodology described in Anderson et al. (2008) and appendix A. Rates of assistance to the agricultural sector are estimated for the half-century since 1955. These are based on estimates of the NRAs for selected individual commodities and RRAs in agriculture and industry. The estimates show that the growth of agricultural protection in Northeast Asia, together with the decline in industrial protection rates, caused RRAs there to rise over the five postwar decades under investigation. The data indicate that agricultural protection rates were negative in the two economies during the low-income stage of development and were low during the lower-middle-income stage, but rose sharply after the economies reached the upper-middle-income stage. The experience seems to be explained by two factors common to rapidly industrializing economies: first, the strong demand for protection among farmers, who shoulder the high cost of intersec- toral adjustments in resource allocations in a context of rapidly declining comparative advantage in agriculture; and, second, the enhanced capacity of nonagriculture to support agriculture. The latter is associated with a growing tolerance among nonfarmers for policies to protect farmers. The first factor emerged at the beginning of the lower-middle-income stage; the second emerged when the economies reached the upper-middle-income stage. Following the sharp increase in agricultural protection during the upper- middle-income stage, the two economies began to suffer from problems arising because of the high level of protection. These included the misallocation of resources and a high fiscal burden in the domestic economy, as well as serious international trade frictions, which are common among high-income economies. These problems might have been less serious if the farm-nonfarm income gap had been addressed properly during the middle-income stage. The agricultural problem in the middle-income stage is known as the disparity problem. The challenge is to balance the conflicting need to support farm incomes and the need to supply low-cost food to a large population of urban workers in view of the weak capacity of the government to raise sufficient revenue for this purpose among nonagricultural sectors. Contrasting patterns in agricultural and industrial growth in Korea and Taiwan, China provide a key to solving this problem. Despite their largely parallel progress in industrial and economic development, the growth in agricultural protection was significantly lower in the latter. The difference seems to be largely explained by differences in industrial and agricultural structures. Industrial structures have been characterized in Taiwan, China by the wide diffusion of small and medium industries in rural areas and, in Korea, by urban-centered 114 Distortions to Agricultural Incentives in Asia industrialization. The cost of intersectoral labor reallocation that had to be shouldered by farmers was therefore considerably lower in the former. The advantage this represented for Taiwan, China was augmented by the success of agricultural diversification there toward highly income-elastic commodities such as horticultural and livestock products and away from traditional subsistence crops, especially rice. The demand for farming assistance was thus significantly less in Taiwan, China despite the pressures of rapid industrial expansion. Devel- oping economies currently pursuing rapid development and relying on a similar strategy of borrowing industrial technology may benefit from closer analysis of the contrasting experiences in Korea and Taiwan, China. Notes 1. Korea and Taiwan, China exceeded US$5,000 per capita GDP in 1983 and 1978, respectively, and US$10,000 per capita GDP in 1991 and 1988, respectively. 2. In terms of real GDP per capita in 2000 constant prices, for example, China exceeded US$1,500 in 1990 and US$5,000 in 2004, whereas Thailand passed the US$1,500 level in 1968 and the US$5,000 level in 1991. In the high-income stage, the United Kingdom and France exceeded US$10,000 in 1960 and 1964, respectively, whereas the United States had already exceeded US$10,000 in 1950. 3. This subsection draws heavily on Moon and Kang (1989). 4. This drove a small wedge between the NRA for producers and the consumer tax equivalent for beef (Anderson 1986). The scheme was similar to one operating in Japan in the 1970s. The reason the government chose the scheme rather than a more efficient, equally protective tariff, plus a consumer subsidy funded with the tariff revenue, is discussed in Hayami (1979) and Anderson (1983a). 5. This section draws heavily on Mao and Schive (1995). 6. Our NRAs for commodities differ from the estimates for Korea in the OECD PSE-CSE Database (2007). There are two key differences. First, our domestic prices are wholesale prices, whereas the data- base relies on farmgate prices. Second, border prices in our calculations are based on the study in Anderson and Hayami (1986), whereas the database relies on a different set of reference prices. The fact that producer prices were above wholesale (consumer) prices in the case of grains and soybeans in Korea is captured by setting the NRA equal to the measured consumer tax equivalent, times the ratio of the NRA to the consumer tax equivalent in Anderson (1989) for the period to 1985 and times the negative of the ratio of producer support estimates to consumer support estimates in the OECD PSE- CSE Database (2007) for the period thereafter. Most of the differences between the database and our measures arise from the differences in the border prices used. For example, our border price for rice in Japan and Korea is the world import unit value, adjusted by a quality coefficient. But the border price for rice in Korea in the database is China's export price for rice, adjusted by transportation costs and, beginning in 2001, the average of the import prices for rice from China, Thailand, and the United States. This means that our NRA series for rice is more stable than the database series in recent years; and, so, this explains the stability of our NRAs for Korean rice compared with the information in the database. For meat products, the border prices are also different. In the estimation of the NRAs for beef, pork, and chicken, the database relies on border prices derived ultimately from data on border prices for meat in Canada or the United States, while we use Japan's import price for beef and own- country unit values for pork and chicken (or, for the 1950s, the import prices in Hong Kong, China). Our approach is preferred for estimating NRAs consistently over longer periods, particularly during the period when Korean imports were absent or negligible. Also, our approach is necessary if one wishes to compare the NRAs of Korea and Taiwan, China on a similar basis. 7. Fertilizer price subsidies were also being provided to farmers in Korea, but they only partially offset the protection from import competition that was provided to the domestic fertilizer Republic of Korea and Taiwan, China 115 manufacturing industry, which meant that farmers were effectively taxed in this respect, too, to support manufacturing (Anderson 1983b). 8. The actual data used to measure the level of agricultural protection in the regression analysis are the nominal protection coefficients (NPC 1 NRP 100). 9. The average RRA in 1955­59 is paired with the agricultural GDP per worker relative to the total GDP per worker in 1955, and so on. 10. The shares of agriculture in GDP and the labor force in Korea were 3.8 and 7.7 percent, respec- tively, in 2004, while the corresponding shares in Taiwan, China in that year were 1.7 and 7.5 percent (see table 2.1, rows 2 and 3). References Amsden, A. 1989. Asia's Next Giant: South Korea and Late Industrialization. New York: Oxford Univer- sity Press. Anderson, K. 1983a. "The Peculiar Rationality of Beef Import Quotas in Japan." American Journal of Agricultural Economics 65 (1): 108­12. ------. 1983b. "Fertilizer Policy in Korea." Journal of Rural Development 6 (1): 43­57. ------. 1986. "The Peculiar Rationality of Beef Import Quotas in Japan and Korea." In The Political Economy of Agricultural Protection: East Asia in International Perspective, ed. K. Anderson and Y. Hayami, chap. 7. London: Allen and Unwin. ------. 1989. "Korea: A Case of Agricultural Protection." In Food Price Policies in Asia, ed. T. Sicular, chap. 4. Ithaca, NY: Cornell University Press. Anderson, K., and Y. Hayami, eds. 1986. The Political Economy of Agricultural Protection: East Asia in International Perspective. London: Allen and Unwin. Anderson, K., Y. Hayami, and M. Honma. 1986. "The Growth of Agricultural Protection." In The Political Economy of Agricultural Protection: East Asia in International Perspective, ed. K. Anderson and Y. Hayami, chap. 2. London: Allen and Unwin. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Anderson, K., and R. Tyers. 1992. "Japanese Rice Policy in the Interwar Period: Some Consequences of Imperial Self Sufficiency." Japan and the World Economy 4 (2): 103­27. Cole, D. C., and Y. C. Park. 1983. Financial Development in Korea, 1945­1978. Cambridge, MA: Harvard University Press. Department of Agriculture and Forestry. Various issues. Taiwan Agricultural Yearbook. Tapei: Depart- ment of Agriculture and Forestry. Douangngeune, B., Y. Hayami, and Y. Godo. 2005. "Education and Natural Resources in Economic Development: Thailand Compared with Japan and Korea." Journal of Asian Economics 16 (2): 179­204. Frank, C. R., K. S. Kim, and K. Westphal. 1975. Foreign Trade Regimes and Economic Development: South Korea. New York: Colombia University Press; Washington, DC: National Bureau of Eco- nomic Research. Gerschenkron, A. 1962. Economic Backwardness in Historical Perspective. Cambridge, MA: Harvard University Press. Hayami, Y. 1979. "Trade Benefits to All: A Design of the Beef Import Liberalization in Japan." American Journal of Agricultural Economics 62 (2): 342­47. ------. 1988. Japanese Agriculture under Siege: The Political Economy of Agricultural Policies. London: Macmillan. ------. 2005. "Emerging Agricultural Problem in High-Performing Economies in Asia." Presidential address at the Fifth Conference of the Asian Society of Agricultural Economists, "Prospects of Asian Agriculture in the New Millennium," University of Sistan and Baluchestan, Zahedan, Iran, August 29­31. 116 Distortions to Agricultural Incentives in Asia Hayami, Y., and M. Aoki, eds. 1998. The Institutional Foundation of East Asian Economic Development. London: Macmillan. Hayami, Y., and Y. Godo. 2004. "The Three Agricultural Problems in the Disequilibrium of World Agri- culture." Asian Journal of Agriculture and Development 1 (1): 3­16. ------. 2005. Development Economics: From the Poverty to the Wealth of Nations, 3rd ed. London: Oxford University Press. Heston, A., R. Summers, and B. Aten. 2006. "Penn World Table Version 6.2." Center for International Comparisons of Production, Income and Prices, University of Pennsylvania. http://pwt .econ.upenn.edu/php_site/pwt_index.php (version of September). Ho, S. P. S. 1979. "Decentralized Industrialization and Rural Development: Evidence from Taiwan." Economic Development and Cultural Change 28 (1): 77­96. ------. 1982. "Economic Development and Rural Industry in South Korea and Taiwan." World Development 10 (11): 973­90. Honma, M., and Y. Hayami. 1986. "Determinants of Agricultural Protection Levels: An Econometric Approach." In The Political Economy of Agricultural Protection: East Asia in International Perspec- tive, ed. K. Anderson and Y. Hayami, chap. 4. London: Allen and Unwin. ------. 2007a. "Distortions to Agricultural Incentives in Korea and Taiwan." Agricultural Distortions Working Paper 30, World Bank, Washington, DC. ------. 2007b. "Distortions to Agricultural Incentives in Japan, Korea, and Taiwan." Agricultural Distortions Working Paper 35, World Bank, Washington, DC. Krueger, A. O., M. Schiff, and A. Valdés, eds. 1991. The Political Economy of Agricultural Pricing Policies. 3 vols. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Kuo, S. W. Y. 1975. "Effects of Land Reform, Agricultural Pricing Policy, and Economic Growth on Multiple Crop Diversification in Taiwan." Philippine Economic Journal 14 (1 and 2). MAF (Ministry for Food, Agriculture, Forestry, and Fisheries, Republic of Korea). various issues. "Major Statistics of Agriculture, Forestry and Fisheries." MAF, Seoul. Mao, Y.-K., and C. Schive. 1995. "Agricultural and Industrial Development in Taiwan." In Agriculture on the Road to Industrialization, ed. J. Mellor, 23­66. Baltimore: Johns Hopkins University Press. Moon, P. Y., and B. S. Kang. 1989. Trade, Exchange Rate, and Agricultural Pricing Policies in the Republic of Korea. World Bank Comparative Study. Washington, DC: World Bank. Myint, H. 1965. The Economic Theory of the Developing Countries. New York: Praeger. OECD PSE-CSE Database (Producer and Consumer Support Estimates, OECD Database 1986­2005). Organisation for Economic Co-operation and Development. http://www.oecd.org/document/ 55/0,3343,en_2649_33727_36956855_1_1_1_1,00.html (accessed December 2007). Schultz, T. W., ed. 1978. Distortions of Agricultural Incentives. Bloomington, IN: Indiana University Press. Scott, M. 1979. "Foreign Trade." In Economic Growth and Structural Change in Taiwan: The Postwar Experience of the Republic of China, ed. J. Galenson, 308­83. Ithaca, NY: Cornell University Press. Tyers, R., and K. Anderson. 1992. Disarray in World Food Markets: A Quantitative Assessment. New York: Cambridge University Press. World Bank. 1993. The East Asian Miracle: Economic Growth and Public Policy. World Bank Policy Research Reports. New York: Oxford University Press. 3 CHINA Jikun Huang, Scott Rozelle, Will Martin, and Yu Liu The purpose of this chapter is to present estimates of indicators of the effects on agriculture in China of direct and indirect government interventions. To put these indicators in context, we review China's experience with policy reforms since the 1950s and measure the extent of these reforms in the agricultural sector. Because of data constraints, we are only able to produce quantitative measures of price dis- tortions since the early 1980s, that is, for the past 25 years, but we also summarize the country's agricultural experience over previous decades. Our review empha- sizes the sectoral and macroeconomic policies and the elements of the institu- tional framework that have influenced the incentive framework facing the sector's product and factor markets. The changes over the past quarter century in trade, pricing, and marketing policies as these affect incentives for producing and con- suming various farm products are reflected in our estimated rates of government assistance to or taxation of producers and consumers. The main finding of our study is that the nature of policy interventions has changed dramatically in China over the past 25 years, propelling agriculture from a sector characterized by significant distortions to one that is relatively liberal. In the 1980s and early 1990s (the early reform period), there were distortions in domestic and external policies that isolated farmers and food consumers from international markets. Importantly, during this period, domestic marketing and pricing policies caused the prices that domestic producers and consumers faced also to become almost independent of the effects of trade policy. Because of this, even in the case of an exportable commodity such as rice, a commodity that appears to have been subject to little distortion at the border (meaning that the international price and the free-market price were nearly identical), domestic 117 118 Distortions to Agricultural Incentives in Asia pricing and marketing policies did not allow producers to reap the profits from international prices. Instead, they forced farmers to sell much of their surplus to the state at an artificially low price. Hence, domestic policies levied a tax on farm- ers although there was little distortion at the border. Similar dynamics character- ized importable commodities such as wheat and soybeans, where, despite fairly high rates of protection erected by trade policies, producers were receiving much less protection than they would have received had there been a free domestic mar- ket for these importables. In contrast, after the 1980s and early 1990s, the liberalization of domestic markets resulted in a reduction in the distortions arising from domestic policies. During this late reform period, the market gradually replaced the government procurement system as the primary mechanism for allocating resources, and mar- ket-generated prices became the basis for production and marketing decisions by farmers. At the same time, especially in the case of importable commodities, trade policy also became more liberal, and the distortions caused by border measures were substantially reduced. As a result, we find that, since the end of the late reform period (that is, after 2000), China's agriculture has become much less dis- torted in two ways. First, the differences between international and domestic mar- ket prices have narrowed considerably for many commodities because of trade policy liberalization. Second, the elimination of domestic policy distortions has meant that, when trade liberalization allows for increased imports or exports of agricultural commodities, prices in the domestic market change and farmer incentives are directly affected. In addition, we also examine the effect of input-oriented policies and exchange rate policies. We find that input-related policies have generated few distortions since 1980. However, exchange rate policy, like changes in pricing, marketing, and trade policies, has played an important role. In the early reform period, exchange rates were highly distorted and served to add an implicit tax on agricultural trade. By the late reform period, however, the system of exchange rates was being reformed gradually, which reinforced the gradual shift toward a more liberal agri- cultural economy. Despite the finding that considerable liberalization has occurred because of policy reforms in the domestic and external economy, there were still distortions in agriculture in the mid-2000s (25 years after the reforms were initiated). In some cases, the remaining distortions have arisen mainly through import tariffs, which, while low by international standards, are still providing a degree of protection for a number of importables (for example, sugar and dairy products). In the case of other importable commodities (for example, maize), the use of export subsidies has created a wedge between the domestic price and the international price. These subsidies are mostly disguised as domestic marketing, transport, and storage China 119 subsidies, which are allowed under World Trade Organization (WTO) rules for developing countries. To show these results and provide the reader with a fuller discussion of recent changes in the structure of China's economy, the policy environment within which the changes have been occurring, and the analytical approach and findings of our analysis of price distortions, we have organized the rest of the chapter as follows. We begin below by examining the changes in the performance of the economy and reviewing the policy environment and reform agenda during peri- ods of change. We examine the changes for two separate periods: the socialist era (1950­79) and the reform era (since 1980). The periods are split in this way not only because of the differences in the performance and the policy changes between the two periods, but also because, in the subsequent distortion analysis, we are only able to quantify the effect of China's liberalizing domestic and external changes after 1980. We discuss our quantitative approach and the sources of our data. The results of the distortion analysis are then presented and discussed. The final section provides concluding remarks. Growth and Structural Change in Agriculture since 1949 The history of the Chinese economy since World War II may be clearly divided into a socialist era that lasted until December 1978 and the era since then during which China has implemented a transition toward a market economy. The socialist era, 1949­78 Socialist policies dominated in China from the 1950s to the 1970s. They had pro- found and complicated effects on agriculture. In this section, we briefly review the performance of the agricultural sector during the socialist era by laying out the sector's successes and failures. We then describe the major policies--inside and outside agriculture--that we believe are responsible for producing the outcomes that were realized. Performance during the socialist era The record of the performance of agriculture in producing food and other raw materials for industry during the socialist era is mixed, although the record partly depends on the standard against which the sector's performance is being judged. Aggregate trends show that agriculture played an important role in increasing food availability, especially in the staple grains (Huang et al. 2007). Between 1952 and 1978, the total area sown only changed marginally, increasing 6.3 percent; the 120 Distortions to Agricultural Incentives in Asia area sown in grain also changed little, declining 2.7 percent. In contrast, grain yields increased by 91 percent from 1952 to 1978, an annual growth rate of 2.8 per- cent. In aggregate, China's grain production rose by 86 percent, a rise of 2.5 per- cent a year. Indeed, the growth rate of grain production outpaced that of the population (1.9 percent), meaning that the agricultural sector increased per capita calorie availability during the socialist era. To the country's credit, the increases in the absolute and per capita levels of availability of food and agricultural raw materials occurred during a time when many other nations were suffering from falling food production. Nonetheless, it is difficult to argue that agriculture's performance was sufficiently stellar to be con- sidered a transformative force in the socialist era economy. Throughout the 1950s, 1960s, and 1970s, China's consumers remained on strictly rationed diets. Coarse grains--maize, millet, and sorghum--and sweet potatoes made up much of the average citizen's staple food intake. Cooking oil, sugar, meat, and vegetables were not available to the typical consumer on a daily basis. Most tellingly, despite the growth, the average level of consumption in the 1970s remained low in urban areas, at only 2,330 calories per capita, while the calorie intake of the average rural resident was only slightly more than the United Nations average minimum requirement of 2,100 calories. Moreover, food production systems were so fragile that they were subject to catastrophic failure, such as occurred during the famine in 1959­61, which reportedly killed more than 30 million people (Ashton et al. 1984). Food availability became such an issue during the late 1960s and the 1970s that China turned to international markets to supplement domestic production. Between 1973 and 1980, China imported an average of more than 6 million tons of grain a year, mostly wheat. In peak import years, grain accounted for a large share of the value of the country's imports. At the time, planners were trying to jump-start industrialization by importing machinery and other technologies. The inability of the agricultural sector to produce sufficient food (and foreign exchange earnings from exports) obviously represented a drag on development. The performance of the food production subsector in the rural economy was mixed, at best. Almost everything else about the record of structural change dur- ing agricultural development in China up to about the late 1970s is negative (Huang et al. 2007). For example, the production structure of cropping showed almost no change at all. In 1952, grain accounted for 88 percent of the total area sown; in 1970, grain was still being sown on 83 percent of the total area sown. Likewise, there was little change in the structure of the agricultural sector accord- ing to a broader definition. The value of the cropping subsector as a share of the value of total agricultural output was 83 percent in 1952; it was still 75 percent in 1970. Perhaps of greatest importance, the income per capita of rural farmers and China 121 other metrics of wealth also reflected the stagnation in the agricultural sector. Despite the rise in grain output, rural earnings per capita in the 1970s were almost the same as they had been in the mid-1950s (Lardy 1983). Even by 1978, nearly 30 years after the start of the socialist era, the annual rural per capita consumption of almost every food item was low, amounting to only 1.1 kilograms of edible oil and 6.4 kilograms of meat (Huang and Bouis 1996). The poverty rate was between 30 and 40 percent. The stagnation in incomes, given the (even modest) rising output, suggests that farm productivity growth was slow. Although the data sources do not facili- tate a rigorous analysis of total factor productivity, there appears to have been a complete absence of any gain in productivity and of any increase in allocative effi- ciency. The work of Stone and Rozelle (1995) and Wen (1993) supports such a conclusion. Using aggregate data, both studies find that total factor productivity growth between 1950 and 1978 was zero or close to zero. Finally, there also was almost no sign of a shift in the structure of employment in the economy. Other rapidly developing countries in East Asia were diversifying the sources of the incomes of rural populations and expanding employment in the off-farm sector, but there was little of this in China's rural sector (Lardy 1983). In 1957, about 84 percent of the population was in the agricultural sector; by 1970, the share had risen to 85 percent; and, in 1980, it was still 83 percent. Of the more than 400 million people in China's rural labor force in 1980, only 4 percent had full-time off-farm employment (deBrauw et al. 2002). In fact, the share of people living and working in the agricultural sector was greater in China than in any other country at a similar level of income (World Bank 1985). Socialist policies and institutions Blame for the poor performance of the agricultural sector may almost certainly be placed squarely on poor policy. Even while local leaders were experimenting with privatized land through ambitious land-to-the-tiller policies in the early 1950s, other factions in the socialist leadership were already developing policies that would threaten the incentives embodied in private landownership (Lardy 1983). The levels of investment believed to be required to promote industrialization were partly obtained through transfers from the agricultural sector. During the plan- ning era, prices for agricultural products were depressed so food could be sold to consumers at rationed prices. The scissors effect was reportedly large (that is, the extent to which the agricultural sector is taxed if the prices of agricultural goods are set below market value and the prices of industrial goods are set above market price). Realized primarily through the direct taxation imposed on the prices of agricultural goods, the estimated taxation rate was 26 percent in 1957 and 27 per- cent in 1978 (Yao 1994). 122 Distortions to Agricultural Incentives in Asia After the early 1950s, farmers were organized into collectives and then com- munes, eliminating the household farm in China. The main negative effect of the communization movement was the absence of incentives. The basic problem was that individual families were not the residual claimants of production, and deci- sion making was left to a collective leadership (Putterman 1993). Farm workers were assigned points based on tasks that were difficult to monitor. There is a debate over the extent to which the collectives were able to motivate farmers to exert effort and attempt to increase the efficiency of production on their farms (Dong and Dow 1993; Chang and Wen 1997; Lin and Yang 2000). Most scholars believe, however, that free riding and the inability to monitor agricultural labor undermined the incentives in agriculture. Socialist era pricing and marketing policies also did little to encourage the effi- cient production and allocation of goods and services. Prices were fixed by the government (Sicular 1988a). Between 1962 and 1978, the price of grain was adjusted only three times and rose by a total of less than 20 percent. Input prices played mainly an accounting function because shortages kept most producers from receiving the quantities they demanded. Marketing institutions--monopolized by government parastatals--did not encourage the development of agriculture. There was little competition, and marketing officials did not have an incentive to search out low-cost or high-quality producers. Through plans directed via the marketing system, production was carried out based on (mostly) planned acreage, target volumes, and the quality and variety of production. Even the ratio between home consumption and marketed surplus was stipulated. The system also served to support--at least in the short run--the govern- ment's effort at forced industrialization by keeping down the price of staples, thereby allowing wages to be kept low. Except for the amount used by farm house- holds for food, feed, and seeds at home, all production of grains, edible oils, and fiber crops was procured exclusively by the state at quota prices for a specified (compulsory) amount (Sicular 1988a). After the early 1960s, the government also procured--at a somewhat higher, above-quota price (to provide an incentive to increase production)--any surplus output beyond the quota and beyond con- sumption by farm families. The incentives, however, were targeted at collective leaders and not at the farmers on whose effort labor depended. To suppress the demand for agricultural products that were in short supply (because they were priced low), marketing policy also exercised tight control over food marketing in urban areas. Almost all major commodities were sold by government agencies to urban consumers and rural households in grain-deficit regions at low prices upon presentation of the appropriate ration coupons. Fertilizers, pesticides, and other material inputs, where available, were also sold during this same period through marketing channels monopolized by the state China 123 (Stone 1988). The Agricultural Inputs Corporation, the government retailer for fertilizer, sold fertilizer to villages on the basis of a carefully formulated plan. Col- lective leaders needed fertilizer coupons to buy fertilizer and other inputs that were in short supply. In an agricultural system dominated by either tens of millions of individual farmers or hundreds of thousands of brigades and teams, there is a need for the government to play a major role in organizing wide-ranging investments because individuals have little incentive to make these investments. In several areas, the gov- ernment made such investments between 1950 and 1978. National leaders arguably put their greatest effort into water conservation (Nickum 1998). In the early 1950s, China's irrigation and flood control systems were a shambles. The irrigated area was less than 20 percent of the total cultivated area (Stone 1988). By 1978, after more than 20 years of investment by local and national governments and the aid of uncountable man-years of corvée labor, more than 40 percent of the country's cropland was being irrigated. By the mid-1970s, every major river system was also protected by intricate networks of dikes, dams, and flood diversion projects. In addition, led by a publicly funded research and development system, China's agricultural scientists led the developing world in many areas and were responsible for generating many scientific breakthroughs. Breeders developed high-yielding semidwarf varieties of rice several years before the Green Revolution began in other parts of Asia (Stone 1988). Farmers were able to use hybrid maize and disease-resistant varieties of wheat in the 1960s and 1970s, long before such technologies were available elsewhere in the developing world. In 1976, Yuan Longping created and commercialized the world's first hybrid rice variety (Lin 1991; Huang and Rozelle 1996). An extension system with nearly 500,000 agents was in charge of introducing the technology to brigade technicians (Hu and Huang 2001). By the mid-1970s, most of the country's cereals were improved varieties. The government's approach to planning, including the placement of the rural economy in the unplanned sector, also had dramatic effects on the nature of employment and on the structural stagnation in the country (Lyons 1987). Because the agricultural sector was large and underdeveloped, leaders decided to make a sharp distinction between people who lived in rural areas and people who lived in urban areas. Agriculture became part of the collective sector. In return for shipments of fertilizer and small amounts of capital and other inputs, the agricul- tural sector was expected to supply food and nonfood commodities to the urban- industrial portion of the economy. The rest of the needs of the collective agricultural sector were supposed to be satisfied by the leadership of the collectives through their own resources. Farmers were not allowed to move freely from their collec- tives. The scope and magnitude of the gap in housing, education, health care, welfare, and other services between rural and urban areas widened throughout 124 Distortions to Agricultural Incentives in Asia the socialist era. Without doubt, these hukou policies--policies tied to the house- hold registration system (hukou)--and other restrictions prevented rural people from moving into manufacturing and service provision. They also artificially lim- ited structural changes during the socialist era and depressed rural incomes and rural productivity. Two key external policies had a negative effect on agriculture in the socialist era. First, in the prereform period, agricultural trade was subject to the planning system (Huang and Chen 1999a, 1999b; Lardy 2001). It was used to supplement the planning for the domestic economy. Given the nation's commitment to self- sufficiency in all areas of the economy, imports were to be used only for procuring those items--most of which were machinery and other productive investments-- that could not be produced domestically and that facilitated realization of the goals of the plan. Almost all trade occurred through eight state-owned trading firms. In the 1970s, the state agricultural trading firms monopolized nearly all food imports and exports. Hence, it was not in the nature of the institutional structure of the trade apparatus to allow specialization in labor-intensive export crops that could be offset by imports of land-intensive staple crops. Agricultural trade was looked upon primarily as a means to generate foreign exchange. Summary: socialist agriculture was a policy-driven disaster After nearly 30 years of development, agriculture was not effective in fulfilling any of its roles. Although output was increasing, this was because of enormous central and local government investments, as well as mostly corvée labor financed pre- dominantly through the sweat of farmers. Productivity and incomes were stag- nant. There had been no structural shift toward more productivity and higher efficiency. A large share of the population was locked into agriculture. The clearest finding arising from our analysis of the socialist era is that the dis- mal performance of agriculture was mainly the result of more than two decades of socialist policies inside and outside the agricultural sector. Production structures, the pricing system, and marketing institutions did not provide adequate incen- tives. Some investments were effective, but there were far too few of these to offset the negative impact of poor incentives. Perhaps most negative was the effects of a system that treated the rural population as second-class citizens and excluded them from a fair share of investments, services, and opportunities. In short, the agricultural and nonagricultural policy environment failed to allow agriculture to contribute to the establishment of a healthy modern economy. Performance during the transition era, post-1978 We first describe the performance of the agricultural sector and the role it has played since the onset of reform in the late 1970s. We then analyze in greater depth China 125 the policy initiatives--inside and outside agriculture--that have helped launch and guide China's agricultural transition. We examine the reform strategy by looking at its various components, including implementation, objectives, and rationale. The ups and downs that characterized the performance of agriculture during the prereform period ceased after 1978. All the best indicators achieved in agricul- tural production from the 1950s to the 1970s were surpassed after the launch of the reforms, and agriculture finally began to play a positive role in the develop- ment process. In the early and mid-1970s, agricultural gross domestic product (GDP) rose by 2.7 percent a year. The annual growth rate almost doubled, to 7.1 percent, during the initial reform period, from 1978 to 1984, before declining to 4.0 percent in 1985­95 and 3.4 percent in 1996­2004 (table 3.1). By world stan- dards, these are high rates of agricultural growth over such a sustained period. At least during the early reform period, output growth--driven by increases in yields--was experienced in all subsectors of agriculture. Between 1978 and 1984, grain production increased, in aggregate, by 4.7 percent a year (table 3.1). Produc- tion rose for each of the major grains, namely, rice, wheat, and maize. The success of agriculture in playing its role of supplying inexpensive food is illustrated by grain prices. During the reforms, with the exception of price spikes in 1988 and 1995, the real prices of rice, wheat, and maize fell by between 33 percent (maize) and 45 percent (wheat) from the late 1970s to the early 2000s. Another change was far more fundamental than the higher yields in grains. As it achieved success in grain production, the agricultural economy was steadily remaking itself from a grain-first economy to an economy producing higher- valued cash crops, horticultural goods, and livestock and aquaculture products. Like the grain subsector, the output of cash crops in general, especially cotton, edible oils, fruits, and vegetables, also grew more rapidly in the early reform period than in the 1970s (table 3.1). Unlike grains, the growth in the nongrain sub- sector continued throughout the reform era (with the exception of land-intensive staples such as cotton). The rise in some subsectors was dramatic. For example, between 1990 and 2004, the increase in vegetable production was so rapid that the country was adding the equivalent of the production capacity of California (the world's most productive vegetable region) every two years. In the total cultivated area in China, the share that is planted in fruit orchards (over 5 percent) is more than double the share in the next closest major agricultural nation. Today, agricul- ture in China is following the dictum "make apples and onions the key link" rather than the dictum "grains first" that was applied to agriculture in the socialist era. Livestock production rose 9.1 percent in the early reform period and has con- tinued to grow at 4.5 to 8.8 percent annually since 1985. The fishery subsector is the most rapidly growing component of food production, growing at more than 10 percent a year between 1985 and 2000. Today, over 70 percent of the world's 126 Distortions to Agricultural Incentives in Asia Table 3.1. Real Average Annual Rates of Economic Growth, China, 1970­2004 (percent) Reform Prereform Indicator 1970­78 1979­84 1985­95 1996­2000 2001­04 GDP 4.9 8.8 9.7 8.2 8.7 Agriculture 2.7 7.1 4.0 3.4 3.4 Industry 6.8 8.2 12.8 9.6 10.6 Services -- 11.6 9.7 8.3 8.3 Population 1.80 1.40 1.37 0.91 0.63 Per capita GDP 3.1 7.4 8.3 7.2 8.1 Grain production 2.8 4.7 1.7 0.03 0.2 Rice Production 2.5 4.5 0.6 0.3 0.9 Area 0.7 0.6 0.6 0.5 1.2 Yield 1.8 5.1 1.2 0.8 0.2 Wheat Production 7.0 8.3 1.9 0.4 1.9 Area 1.7 0.0 0.1 1.4 5.1 Yield 5.2 8.3 1.8 1.0 3.3 Maize Production 7.4 3.7 4.7 0.1 5.5 Area 3.1 1.6 1.7 0.8 2.5 Yield 4.2 5.4 2.9 0.9 2.8 Other production Cotton 0.4 19.3 0.3 1.9 6.5 Soybeans 2.3 5.2 2.8 2.6 2.4 Other oil crops 2.1 14.9 4.4 5.6 0.6 Fruit 6.6 7.2 12.7 8.6 29.5 Meat 4.4 9.1 8.8 6.5 4.6 Fishery 5.0 7.9 13.7 10.2 3.5 Planted area Vegetables 2.4 5.4 6.8 6.8 3.8 Fruits 8.1 4.5 10.4 1.5 2.2 Sources: Data of the National Bureau of Statistics and the Ministry of Agriculture. Note: Growth rates are computed using the regression method. The growth rates of individual commodities and groups of commodities are based on production data. GDP in real terms in 1970­78 is the growth rate of national income in real terms. -- no data are available. freshwater aquaculture is produced in China. The sustained and rapid rise in live- stock and fishery output has steadily eroded the predominance of cropping. After remaining fairly static during the socialist era, the share of agriculture contributed by cropping fell from 76 to 51 percent between 1985 and 2005. Meanwhile, the China 127 Table 3.2. Structure of the Agricultural Economy, China, 1970­2005 (share in agricultural output, %) Subsector 1970 1980 1985 1990 1995 2000 2005 Crop 82 80 76 65 58 56 51 Livestock 14 18 22 26 30 30 35 Fisheries 2 2 3 5 8 11 10 Forestry 2 4 5 4 3 4 4 Source: Data of the National Bureau of Statistics of China. combined share of livestock and fisheries rose to 45 percent, more than doubling the 1980 share of 20 percent. The expansion in the livestock and fishery subsectors is also outpacing the growth in most of the cropping subcategories (tables 3.1 and 3.2). It is projected that, beginning in 2008, cropping will account for less than 50 percent of agricultural output in China. Moving off the farm The reform era has also generated other fundamental transformative changes. One sees this by examining the rural economy based on a definition that is broader than agriculture. While the annual growth in agriculture averaged about 5 percent throughout the reform era, the growth rates of the economy as a whole and in the industrial and services sectors were more rapid (table 3.1). In fact, since 1985, the growth in industry and services has been two to three times more rapid than the growth in agriculture. Because of the differences in the sectoral growth rates, agriculture's share of GDP fell from 40 percent in 1970 to less than 13 per- cent in 2005 (data of the National Bureau of Statistics). Given the current growth of agricultural GDP and overall GDP, the share of agriculture will fall below 10 percent after 2010. The shifts in the economy may also be seen in employment. Agriculture employed 81 percent of the labor force in 1970. By 2005, however, as the industrial and services sectors grew in importance, the share of employment in agriculture fell to a reported 45 percent. Agricultural trade liberalization While much has been made of China's accession to the WTO as a turning point in its relationship with the world, the country's open-door policy was undertaken much earlier (Huang and Chen 1999a, 1999b). During the process, China has abandoned isolation and become one of the world's great trading nations, includ- ing in agricultural trade. From 1980 to 2000, the total value of the country's agri- cultural trade grew by about 6.0 percent a year. It has more than doubled since 2000, making China the fourth largest importer of agricultural commodities in 128 Distortions to Agricultural Incentives in Asia Figure 3.1. Agricultural Trade Balance, by Factor Intensity, China, 1985­2002 12,000 10,000 8,000 6,000 millions 4,000 US$ 2,000 0 2,000 exports, 4,000 net 6,000 8,000 10,000 1985 1987 1989 1991 1993 1995 1997 1999 2001 year labor-intensive commodities land-intensive commodities Source: Rosen, Huang, and Rozelle 2004. Note: Labor-intensive commodities include fruits, vegetables, meat products, and aquaculture products. Land-intensive commodities include food and feed grains, soybeans, edible oils, and cotton. the world (Gale 2006). In agriculture, the level of exports has exceeded that of imports almost every year since the reforms were launched (Huang and Chen 1999a, 1999b). Perhaps more remarkable is the shift in the composition of trade over the past 25 years. Figure 3.1 shows that net exports of land-intensive bulk commodities, such as grains, oilseeds, and sugar, have fallen, while exports of higher-valued, more labor-intensive products, such as horticultural, livestock, and aquaculture products, have risen. The country has thus begun to export those commodities in which it has a comparative advantage and import those in which it does not have an advantage. Disaggregated, crop-specific trade trends show the same sharp shifts (Anderson, Huang, and Ianchovichina 2004). The production and marketing environment After more than 25 years of reform, one of the most striking changes in the nature of agriculture in China is the change in the role of local and national government regulation in production and marketing. During the socialist era, bureaucrats in government supply and marketing agencies and local commune and brigade offi- cials were deeply involved in all aspects of pre- and postharvest decision making. By 2005, the situation had changed dramatically. Indeed, a notable feature of China 129 China's agricultural economy today is the limited extent of government interven- tion (with several exceptions). Restrictions on landownership aside, agriculture in China may now be one of the least regulated domestic agricultural economies in the world. In a recent survey carried out by the Center for Chinese Agricultural Policy, all farmers who were not renting village-owned orchards planted in the early reform period indicated that they had made the planting decision and had not been compelled by local officials (Rozelle, Huang, and Sumner 2006). All farmers in a survey of households in eight provinces stated that they had pur- chased all their chemical fertilizers on their own and that local officials had no role in the transactions (Zhang, Li, and Rozelle 2005). Moreover, all purchases had been made through private suppliers. On the procurement side, government parastatals used to be responsible for purchasing farm output, but, today, a large majority of the grains, oilseeds, and fiber crops and all the horticultural and livestock products are sold to small pri- vate traders (H. Wang et al. 2006). Indeed, despite the impressive rise of super- markets and processing firms catering to the retail needs of urban populations, a recent survey found that almost all purchases of fruit, vegetables, nuts, livestock, and other agricultural products are made by individual entrepreneurs who are trading on their own account. The existence of millions of small traders who are competing in a market virtually without regulations has meant that the market has become well integrated and efficient (Park et al. 2002; Huang, Rozelle, and Chang 2004; Rozelle and Huang 2004, 2005). Productivity trends and rural incomes While it is possible that trends in agricultural productivity tell a story about the effects of transition on agricultural performance that is somewhat different from the story told by output (as they did during the prereform period), this is not the case. First, as shown in table 3.1, output per unit of land rose sharply (that is, all yields). In addition, trends in agricultural labor productivity--meas- ured as output per farm worker--paralleled those in yields during the entire reform period. Moreover, several series of estimates of total factor productivity have been produced for China's agriculture (McMillan, Whalley, and Zhu 1989; Fan 1991; Lin 1992; Wen 1993; Huang and Rozelle 1996; Fan 1997; Jin et al. 2002). The studies uniformly demonstrate that, in the first years after reform (1978­84), comprehensive measures of productivity (either constructed total factor productivity indexes or their regression-based equivalents) rose 5 to 10 percent a year. Rural incomes during the reforms steadily increased partly because of rising productivity and, perhaps even more, because of the increasing efficiency associ- ated with specialization, the shift to more production of higher-value crops and 130 Distortions to Agricultural Incentives in Asia Table 3.3. Rural Income per Capita, China, 1980­2001 (real 2000 yuan) Annual growth, Income group 1980 1985 1990 1995 2000 2001 1980­2001, % Average 711 1,248 1,305 1,702 2,253 2,347 6 Bottom decile 312 448 442 493 579 578 3 (poorest) Top decile 1,530 2,486 3,253 4,763 6,805 7,159 8 (richest) Source: Compiled based on data of the National Bureau of Statistics. livestock commodities, and the expansion of off-farm work (table 3.3). Between 1980 and 2001, average real rural per capita income rose by a remarkable 6 percent per year, which is as high as the growth rates experienced in Japan and the Republic of Korea during the take-off years of those economies. The amount of attention given to the rural income problem by the media might seem surprising, but the attention has been generated by the more rapid rise in urban incomes relative to rural incomes. Inequality is increasing between rural and urban areas, but it is also increasing between those people within the rural economy who were relatively rich at the outset of reform and those who were relatively poor. The growth rate in rural per capita income in the richest decile, at more than 8 percent annually, is far higher than the average (3 percent annually), which means that the growth rate among other segments of the rural population is correspondingly lower than 3 percent. In relative terms, the poorest of the rural poor are falling behind. Summary of agriculture's performance during the transition era Whereas the socialist era witnessed few transformations, the agricultural sector in China has changed dramatically during the transition. The sector has grown, but the decline in its importance in employment and in the value of output in the overall economy has been characteristic of the growth. The structure of the sec- tor itself is also changing; it is diversifying out of coarse grains into fine grains, out of staple grains into higher-valued crops, and out of cropping into livestock and aquaculture. Trade patterns are changing and are now more reflective of the country's comparative advantages. One of the largest shifts has occurred in pro- duction and marketing, which have become almost laissez-faire, that is, there is little government intervention. Although the most dramatic changes have taken place most rapidly among wealthier households, changes are also occurring among the poor. China 131 Institutions and the Policy Foundation of Reform Unlike leaders in the transitional economies in Europe, leaders in China did not seek to dismantle the planned economy in favor of liberalized markets during the initial stages of reform (Rozelle and Swinnen 2004). Policy makers in China only began to shift their focus toward market liberalization in 1985 after decollectiviza- tion was complete. Even then, liberalization was start-stop (Sicular 1995). Lin, Cai, and Li (1996) argue that the leaders were afraid of disruptive social effects. The institutions through which the leadership controlled the main goods in the food economy (such as grains, fertilizers, and meat products) could not be elimi- nated before the institutions were in place that would support more efficient mar- ket exchanges. Throughout the reforms, leaders were investing in the economy and changing the rules governing those domestic producers and consumers who were interacting with the external economy. Pricing policies Early during the reforms, although China's leaders had no concrete plan to liber- alize markets, they did take steps to change the producer incentives that were embodied in the prices producers received for their marketed surplus. Perhaps one of the least appreciated initiatives of the early reformers was their bold deci- sion to increase the administered prices received by farmers (Lardy 1983; Sicular 1988a). Between 1978 and 1983, in a number of separate actions, the government- set above-quota prices (the payments farmers received for voluntary sales beyond the mandatory deliveries) were increased by 41 percent for grain and by around 50 percent for cash crops (Sicular 1988a). According to National Bureau of Statis- tics data, the price of grain relative to fertilizer rose by more than 60 percent dur- ing the first three years after the reform. During the early reform period, the rise in the above-quota price represented a higher output price at the margin for farmers (Sicular 1995). The important contribution of pricing policy was in the timing and breadth of the policy change. The first major price rise occurred in 1979, close to the moment when reformers were deciding to decollectivize. However, following the decision of the leadership to implement the household responsibility system gradually, beginning first in the poorest areas of China, the price increases immediately affected all farmers, regardless of whether they had been decollectivized. By 1981, the period of the second major price increase, less than half of all farmers had been allowed to dismantle their communes (Lin 1992). Hence, as long as there was a link, albeit weak, between the output price and production, the plan-based price 132 Distortions to Agricultural Incentives in Asia rise would have led to increases in farm output. Empirical studies confirm the strong impact of these price changes on output during the first years of transition (Fan 1991; Lin 1992; Huang and Rozelle 1996; Fan and Pardey 1997). Greater incentives The rural economic reforms initiated in 1979 were founded on the household responsibility system. The reforms of this system led to the dismantlement of the communes. Agricultural land was contracted to households, mostly on the basis of family size and the number of people in the household labor force. Most importantly, after the reforms of the system, income and control rights over farm- land belonged to individual households (though the right to sell was excluded). There is little doubt that the changes in incentives resulting from the reforms in property rights triggered strong growth in farm output and productivity. In the most definitive study on the subject, Lin (1992) estimates that China's household responsibility system accounted for 42 to 46 percent of the total rise in output during 1978­84. Fan (1991) and Huang and Rozelle (1996) find that, even accounting for technological change, institutional change during the late 1970s and early 1980s contributed about 30 percent of output growth. Empirical researchers have also documented impacts that go beyond output. For example, McMillan, Whalley, and Zhu (1989) demonstrate that the early reforms raised total factor productivity, accounting for 90 percent of the rise (23 percent) in 1978­84. Jin et al. (2002) show that the reforms had a large effect on productivity, contributing substantially to a rise in total factor productivity that exceeded 7 per- cent annually. Domestic market liberalization policies In addition to pricing changes and decollectivization, another major task of the reformers was to create efficient institutions of exchange. Whether they are classic and competitive or only workable substitutes, efficient markets facilitate transac- tions among agents to allow specialization and trade and, through pricing mecha- nisms, provide information to producers and consumers about the relative scarcity of resources and products. The major changes to agricultural commerce in the early 1980s almost exclu- sively centered on increasing crop purchase prices (Sicular 1988a; Watson 1994). The decision to raise prices should not be considered part of a strategy to liberal- ize markets. Planners in the Ministry of Commerce made the price changes administratively, and the changes were mostly executed by the national network of grain procurement stations under the direction of the State Grain Bureau. China 133 An examination of policies and marketing activities in the early 1980s reveals the limited changes in the marketing environment in China's food economy before 1985. It is true that reformers allowed farmers greater discretion in produc- ing and marketing crops in 10 planning categories, including vegetables, fruits, and coarse grains. Moreover, by 1984, the state only claimed control over 12 com- modities, including rice, wheat, maize, soybeans, peanuts, rapeseeds, and several other cash crops (Sicular 1988a). While this may seem to represent a significant step toward liberalization, the crops remaining almost entirely under the planning authority of the government still accounted for more than 95 percent of the total sown area in 1984. Hence, in policy and practice, the output of almost all the sown area and the related marketing decisions were still directly influenced by govern- ment planners. The planners proceeded with equal caution in reducing the restrictions on free-market trade, and the process was start-stop-start at first. Market liberalization began in earnest after 1985. Changes in the procurement system, additional reductions in the restrictions on commodity trading, steps to commercialize the state trading system for grains, and calls for expansion in the establishment of markets in rural and urban areas led to a surge in market- oriented activity (Sicular 1995). For example, only around 240,000 private and semi- private trading enterprises were registered with the State Markets Bureau in 1980; by 1990, there were more than 5.2 million (deBrauw, Huang, and Rozelle 2004). Private trading gradually expanded to include all categories of agricultural out- put remaining after contractual obligations to the government had been fulfilled. This eventually undermined the state marketing system (Rozelle et al. 2000). The reformers eliminated the planned procurement of all agricultural products except rice, wheat, maize, and cotton. Government commercial departments were able to continue to buy and sell only if they did so through the market. Incentives were introduced for grains through reductions in the volume of compulsory quotas and increases in procurement prices. Even among grains, after the share of com- pulsory quota procurement reached 29 percent in 1984, it declined to 18 percent in 1985 and 13 percent in 1990. The share of negotiated procurement at market prices rose from 3 percent in 1985 to 6 percent in 1985 and 12 percent in 1990. Agricultural technology and water infrastructure development Plant breeding and agricultural research are almost universally organized and managed by the government. Reflecting the urban bias of food policy, most crop breeding programs emphasize fine grains (rice and wheat). The traditional focus of research has been on quantity because of concerns about food self-sufficiency, but quality improvement has recently been emphasized because of a shift in demand toward higher-quality food. There have been several joint ventures and 134 Distortions to Agricultural Incentives in Asia private domestic investment initiatives in agricultural research and development, but government policy still discriminates against the practice. The record of the reforms in agricultural technology is mixed, and the impact of the reforms in terms of new technological development and greater crop pro- ductivity is unclear. Empirical evidence indicates that the effectiveness of the country's agricultural research system is declining (Jin et al. 2002). Our previous work has shown that, while competitive grant programs probably boosted the effectiveness of the system, the reliance on revenues from commercialization to make up for falling budgetary commitments weakened the system.1 It is possible that imperfections in the seed industry contributed to the lack of success in the reforms in research on crop breeding. Since the late 1990s, there has been a sharp rise in spending on agricultural research and development. However, government investments in the development of water management for irrigation and flood control swamp the investments in agricultural research. From the 1950s to the 1970s, most of the government's efforts were focused on building dams and canals, often with the input by farmers of corvée labor. Since the end of the 1970s, greater attention has been focused on increasing the use of the country's massive groundwater resources (Wang, Huang, and Rozelle 2005). By 2005, China had more tube wells than any other country in the world, with the possible exception of India. Initially, the investments were made by local governments, which were assisted by county and provincial water bureaus. But, by the 1990s, the government was encouraging the huge shift in ownership that was already occurring, as pump sets, wells, and other irrigation equipment were being taken over by families on private farms (J. Wang et al. 2005). Meanwhile, private water markets--whereby farmers would sell water pumped from their own wells to other farmers in the village--were also encour- aged. The main policy initiative after the mid-1990s in the surface water subsector was a management reform that sought to raise efficiency in water use. Trade policy There were important changes during the 1980s and 1990s in foreign exchange policy that saw the nation's currency depreciate steeply and trading rights become more accessible to traders. A number of other fundamental reforms were also undertaken in China's international trading system. The lower agricultural tariffs and the rising imports and exports of agricultural products began to affect the domestic terms of trade in the 1980s. During the initial reform years, most of the fall in protection was caused by a reduction in the commodities that were con- trolled by single-desk state traders (Huang and Chen 1999a, 1999b). In many products, competition among nonstate foreign trade corporations started to China 135 stimulate imports and exports (Martin 2002). Although many major agricultural commodities were not included in the effort to decentralize trade, the effort spurred the export of many agricultural goods. In addition, policy shifts in the 1980s and 1990s also altered the trading behavior of state traders. Thus, the policy leadership allowed state traders to boost imports during these decades. Steps to relax restrictions on the right of access to import and export markets were matched by steps to reduce the taxes assessed at the border. After the elimi- nation of limits on the imports and exports of many agricultural commodities, a new effort was undertaken in the early 1990s to lower the level of formal protec- tion. The simple average agricultural import tariff fell from 42 percent in 1992 to 24 percent in 1998 and 21 percent in 2001 (Rosen, Huang, and Rozelle 2004). Overall protection in the agricultural sector has declined in the past 20 years. Much of the drop has been caused by the decentralization of authority over imports and exports, a relaxation in the licensing procedures for some crops (for example, the shift away from state trading firms in importing oil and oilseeds), and changes in the foreign exchange rate. Other trade policies have reduced the scope of nontariff barriers, relaxed the real tariff rates at the border, and changed quotas (Huang and Chen 1999a, 1999b). Despite these real and, in some areas, rapid reforms, the control over commodities that policy leaders consider of national strategic importance, such as rice, wheat, and maize, remains largely with government officials (Nyberg and Rozelle 1999). The accession to the WTO has been a major event in the country and will con- tinue to have a substantial impact on many sectors. Nonetheless, given the trends in reform, the WTO accession agreement in the agricultural sector, in its most basic terms, represents a continuation of previous agricultural policies. Importantly, though, the commitments embodied in this agricultural part of the agreement-- market access, domestic support, and export subsidies--are a binding confirma- tion and codification of steps the government took in the 1990s while it was negotiating the accession to the WTO. Summary: China's transition era agricultural policies The scope of China's policy efforts during the transition is impressive. Policy shifts were realized in pricing, production management, marketing, investments, tech- nology, and trade. Although the rate of investment has risen during the reforms, the country is probably still underinvesting in agriculture relative to other sectors and other countries. Explicit taxes and the taxes implicit in pricing and trade poli- cies have also fallen. The government is not assisting the agricultural economy in the substantial ways that characterize the country's neighbors in East Asia (see Honma and Hayami 2008). However, it appears to have headed in the direction 136 Distortions to Agricultural Incentives in Asia noted by Timmer (1998), whereby developing nations, after they reach a certain point in development, begin to shift from making net extractions from agriculture to making net investments in the sector. Many policies and other factors outside agriculture have affected the sector. Other rural policies, such as those that govern fiscal reform, the emergence of township and village enterprises, privatization, and rural governance, almost cer- tainly have a large, albeit indirect effect on agriculture. Urban employment poli- cies, residency restrictions, exchange rate management, and many other policy initiatives also affect agriculture by influencing relative prices in the economy, access to jobs off the farm, and the overall attractiveness of staying on the farm. Taken together, these policies have had a dramatic impact on the agricultural sector. They have increased the output of food, driven prices down, and improved the supplies of nongrain food and raw materials for industry. The mix of policies--pricing, enhanced property rights, market liberalization, investment, trade--has also made producers more efficient. It has freed up their labor and resources, and these have supported the structural transformation in the agricul- tural economy and the rural economy more generally. There are several pieces of convincing evidence that agriculture is beginning to play a more effective role in the nation's development: the importance of grain in the cropping subsector is shrinking; the importance of the cropping subsector in the overall agricultural sector is shrinking; and the importance of agriculture in the general economy is shrinking. Productivity is up, and rural incomes are up. Many of the increases in welfare, however, are being generated by individuals--there have been more than 200 million of them--who have been able to shift from grain production to high- valued crops or from cropping to livestock and fisheries production or, perhaps most importantly, who have been able to leave agriculture and the rural economy altogether and obtain employment in cities. Quantifying the Distortions to Agricultural Incentives since 1980 We introduce our quantification of the policy-imposed distortions in the incen- tives faced by farmers in China by explaining our methodology and, in particular, the way we deal with distortions in the market for foreign currency. We also out- line our data sources. We then discuss our results in detail. Methodology and data sources The main focus of our study is government-imposed distortions that create a gap between domestic prices and prices as they would be under free markets. It is not China 137 possible to understand the characteristics of agricultural development through a sectoral view alone. For comparative evaluation, the project's methodology there- fore involves estimates of the effects of direct agricultural policy measures, includ- ing distortions in the foreign exchange market, but also estimates of distortions in nonagricultural sectors. Specifically, we compute nominal rates of assistance (NRAs) to farmers that reflect an adjustment for direct interventions on tradable and nontradable inputs. We also generate NRAs for nonagricultural tradables for comparison with the NRAs for agricultural tradables via the calculation of a rela- tive rate of assistance (RRA) (see Anderson et al. 2008 and appendix A). This approach is not well suited to the analysis of the prereform period because prices played only an accounting function at the time, and key prices involving currency exchange rates were enormously distorted. During the reform period, however, the price comparison approach has provided valuable indicators of distor- tions to the incentives for production, consumption, and trade. as well as indicators of the income transfers associated with policy interventions. Such an approach is necessary because of the low inframarginal procurement prices, but also because of the complexity and lack of transparency in the trade barriers in agriculture, includ- ing tariffs, licenses, quotas, tariff rate quotas, and government trading. Exchange rate distortions present particular measurement problems and require detailed analysis if we expect measures based on price comparisons not to be misleading. NRAs and CTEs NRAs are constructed in three steps. The NRAs on agricultural output are based partly on a comparison of the price of a commodity in the domestic economy at the port and the international price of the commodity at that same border point (that is, including, cost, insurance, and freight in port for an import good and free on board in port for an exportable good), taking into account inherent differences in product quality, where appropriate. Conceptually, we need this part of the NRA on output to measure the extent, at the border, of the distortions caused by tariffs, exchange rate policies, and other nontariff barriers. A positive NRA measure indi- cates that the sector is being protected from import competition, while a negative NRA measure indicates the sector is subject to some form of (possibly implicit) taxation at the border. Because we have independent observations on the prices obtained by farmers in local markets, the second step involves estimating NRAs on output at the farm- gate by also taking into account any domestic distortions affecting the returns to farmers. This part of the NRA is calculated after allowing for quality adjustments, taxes, subsidies, and the costs of domestic transport, storage, and handling in delivering products from the farm to the wholesale market. This part of the NRA on output arises from subsidy or transfer payments that cause the prices received 138 Distortions to Agricultural Incentives in Asia by farmers to differ from the prices they would receive under competitive internal market conditions. As we show below, these internal subsidies and tax measures had an enormous impact on the returns to farmers, particularly during the early reform period. An important component of this second part of the NRA evolved during the period when domestic incentives were distorted by the requirement to deliver a fixed quota of output at a (low) planned price, although producers were allowed to sell above-quota output (surplus) at a higher market price. Clearly, the delivery requirements reduced the incomes of farm households, but the impact on the level of output is less clear. Sicular (1988b) makes the point that, in the short run, the out- put decision of a firm is determined by the marginal return rather than the average return. Under these circumstances, the inframarginal tax imposed by the plan deliv- ery requirement may not have reduced the incentive to produce. However, more recent work on decoupled price incentives, by taking into account the incentives for firm entry and exit, suggests that average returns may actually have a greater impact on output in the longer term (de Gorter, Just, and Kropp 2008). Nonetheless, the inability of farm households to sell their land probably reduced the extent to which these households exited from agriculture in response to reductions in incomes. The third step in our NRA measurement captures distortions in farm input markets. To realize this step, we provide measures that take into account direct subsidies and differences between international input prices and the input prices that farmers pay. While these forms of protection (or taxation) may be important in other countries, we find that they are generally relatively small in China and so contribute little to the overall NRA. While most of our focus is on agricultural producers, we also consider the extent to which consumers are taxed or subsidized. To achieve this, we use a meas- ure called the consumer tax equivalent (CTE). In principle, this measure com- pares the price that consumers pay for food commodities and the international price at the border. As with the NRAs, differences arise between NRAs and CTEs because of distortions in the domestic economy, as well as at the border. The domestic distortions are caused by transfer policies, taxes, and subsidies that gen- erate gaps between the prices paid by consumers (adjusted to the wholesale level) and domestic market prices. If a CTE is negative, then consumers are paying prices that are below the international market prices and the consumers are being subsidized. (For more details on the methodology used to calculate these meas- ures, see Anderson et al. 2008 and appendix A.) The foreign exchange regime Prior to 1981, China's official exchange rate was seriously overvalued. While this did not directly affect exports and imports because decisions on these were made China 139 by planners, it did create serious accounting difficulties given that exports gener- ally incurred losses (Lardy 1992). If the official exchange rate is used, it will pro- vide misleading indicators of the incentives created by the foreign exchange regime. Because it made all foreign goods appear inexpensive in terms of the domestic currency, it overestimated the extent of protection provided to any importable good. In 1981, an internal settlement rate was introduced that was intended to be aligned with the average cost of foreign exchange. This supplied at least some basis for meaningful comparisons between domestic and international prices. The internal settlement rate, at 2.8 yuan per U.S. dollar for trade transactions in 1981, represented nearly a 50 percent devaluation relative to the official exchange rate. It was used only for nontrade transactions. This internal rate remained at Y 2.8 until January 1985, when it was merged with the official exchange rate. During most of the reform period, the effects of the foreign exchange regime were relatively transparent. The regime between the late 1970s and 1994 was included among those found by Kiguel and O'Connell (1995) to involve differen- tial rates for different types of current account transactions. The overvalued offi- cial exchange rate was a key element of this system. Prior to 1979, enterprises had to surrender all their foreign exchange earnings at the official rate. However, beginning in that year, exporting enterprises were allowed to retain some of their foreign exchange earnings (Lardy 1992). Because of the pervasive shortage of for- eign exchange in the economy, the value of these retained earnings was, on aver- age, considerably above the value as measured by the official exchange rate, although there were restrictions on trading retained foreign exchange among enterprises, which inevitably exposed the variation in the demand of enterprises for foreign exchange. The restrictions on trade in retained foreign exchange were gradually liberal- ized, and a legal secondary-market exchange rate emerged that was higher (more depreciated) than the official exchange rate. There was always a shortage of for- eign exchange at the official exchange rate, forcing importers to meet their needs for additional foreign exchange at the secondary-market rate. Under these cir- cumstances, the exchange rate system created a distortion analogous to a tariff and an export tax. The exchange rate received by exporters was lower than that paid, at the margin, by importers. To account for the effects of the exchange rate system, we construct an exporter exchange rate series by using the retention ratio to calculate a weighted average of the official and the secondary-market exchange rates. We use the secondary- market exchange rate as an indicator of the price paid for foreign exchange, at the margin, by importers. Following the methodology outlined in Anderson et al. (2008) and appendix A, we calculate an estimated equilibrium exchange rate as 140 Distortions to Agricultural Incentives in Asia the simple average of the importer and the exporter exchange rates. This method arbitrarily assumes equal elasticities of demand and supply for foreign exchange, though this assumption does not affect the results for any trade impacts because taxes on imports and taxes on exports have the same effect (the Lerner symmetry theorem). The share of export earnings eligible to be retained rose gradually, and the extent to which these eligibility rights were tradable increased. Initially, there were only limited opportunities to trade the rights, and the shadow value varied from firm to firm depending upon the firm's foreign exchange earnings and need for foreign exchange. Policy makers quickly recognized that it was important to be able to able to transfer foreign exchange among firms. Lardy (1992) notes that for- eign exchange trading outlets were first established in Guangzhou (1980) and Shanghai (1981). However, the exchange rates in these markets tended to be heav- ily managed; the government was seeking to set the selling price at, for example, the internal settlement rate (Lardy 1992). Formal foreign exchange adjustment centers were established on a large scale after 1985. The centers allowed firms with excess foreign exchange earnings to sell to ventures that sold their output domestically and needed foreign exchange. Over the next few years, a large network of these centers was established, and the mar- kets became more closely integrated over time. By 1988, Lardy (1992) concludes, the price in the Shanghai market followed the strictures of supply and demand, subject to the conditions on the use of foreign exchange, including licensing requirements and duties on imported goods. The pricing policies for imported goods were also becoming more liberal, and, by 1990, the prices of close to 90 per- cent of all imported goods were based on the import price, plus transport costs and tariffs (Lardy 1992). During the transition, the exchange rate regime had an extremely important influence on the returns obtainable through exported goods and on the prices paid for imported goods. Even in the absence of explicit trade policies, overvalued official exchange rates tended to lower the returns to exported goods and increase the costs of imported goods, often by large amounts. The combination of an overvalued offi- cial exchange rate and a secondary-market exchange rate at which importers could purchase foreign exchange legally allows us to assess the effects of the foreign exchange regime and then to begin to assess the impacts of other trade policies. The analytics of a multiple exchange rate system are relatively clear and are eas- ily understood in a partial equilibrium setting. Figure 3.2 shows the exchange rates in a market that is characterized by an upward-sloping supply of foreign exchange (perhaps determined by the marginal costs of generating additional assets) and a downward-sloping demand for foreign exchange (perhaps determined by the extent of substitutability among imports and domestic goods). If there is an official China 141 Figure 3.2. The Domestic Market for Foreign Currency, China Sfx Em currency Em E foreign Es of E0 Dfx price QS QS QE quantity of foreign currency Source: Anderson et al. 2008; see appendix A. Note: See the text for an explanation of the figure. Sfx the supply of foreign exchange. Dfx the demand for foreign exchange. exchange rate, E0, at which exporters must surrender their foreign currency to the central bank and a secondary-market exchange rate, Em, at which importers may buy foreign exchange, then the two-tier exchange rate system functions as a uni- form tax on all exports or (equivalently) a uniform tax on all imports. Figure 3.2 shows the effect of such a regime. Setting the official exchange rate at E0 reduces the returns to exporters relative to the equilibrium rate, E. The resulting shortage of foreign exchange drives up the foreign exchange's scarcity value, and, in the pres- ence of a secondary market, its market price in sales to importers becomes Em. Under these circumstances, the two-tier exchange rate reduces export earnings from QE to QS. A foreign exchange retention scheme of the type used in China during the tran- sition raises the return to exports by allowing exporters to convert some of their foreign exchange earnings at the higher secondary-market rate. As a result, the sup- ply of foreign exchange rises from QS to QS . The increase in supply allows the price of foreign exchange on the secondary market to fall to Em . This reduces the cost of imported goods, and it also raises the demand for imports from QS to QS . The tax on exporters was lowered because exporters were allowed to retain a portion of their foreign exchange earnings and sell foreign exchange on the sec- ondary market. The retention rates have been estimated roughly at 20 percent in 1981­84, 25 percent in 1985­86, 44 percent in 1987­90, and 80 percent in 1991­94 (World Bank 1994). The resulting blended average exchange rate received by exporters is indicated in the figure as Es . The introduction of the for- eign exchange retention scheme reduced the secondary-market rate because of the increased incentive to supply foreign exchange. 142 Distortions to Agricultural Incentives in Asia We use several different series to obtain secondary-market exchange rate data: the internal settlement rate in 1981­84, an estimated secondary-market exchange rate in 1985­86, and the Foreign Exchange Adjustment Center rate in 1987­94. We use the difference between the importer exchange rate and the estimated equilibrium rate as a measure of the component of the protection for import- competing products that is represented by exchange rate distortion. Similarly, we use the difference between the exporter exchange rate and the equilibrium rate as a measure of the exchange rate distortion affecting exportable goods. Relying on these principles and data, we obtain the results described in Huang et al. (2007).2 The data used In compiling price data, we first had to make choices about our study's coverage of commodities. We have included 11 commodities: rice, wheat, maize, soybeans, cotton, pig meat, milk, poultry, fruit (we use apples as a representative product), vegetables (we use tomatoes as a representative product), and sugar (sugar beets and sugarcane). Over the period we study, these commodities account for between 75 percent (in the late 1980s) and 60 percent (during the early 2000s) of the total value of agricultural output (Huang et al. 2007). Because decisions on production and consumption were only gradually being allowed to respond to domestic market prices and because we do not have access to reliable data on sec- ondary-market exchange rates prior to 1981, we focus on data for the period beginning in 1981. The data in our study have been gathered from diverse sources, depending on the period of analysis and the commodity (see the references). We have taken com- modity balance data (production, utilization, trade, and others) from the CAPSiM database of the Center for Chinese Agricultural Policy. These data have been com- piled mainly by the Ministry of Agriculture (production), the National Bureau of Statistics (consumption and others), and the Ministry of Commerce (trade). The data on domestic prices are from several ministries. Specifically, farmgate output prices have been taken from the cost of production surveys conducted by the National Development and Reform Commission. Data on the wholesale and retail prices of most products have been collected from the Center for Price Monitoring, the National Development and Reform Commission, the Ministry of Agriculture, and the Department of Rural Survey at the National Bureau of Statistics. If whole- sale and retail prices for some commodities are not available for some years, we have estimated the price margins from farmgate to wholesale and to retail. Much of the data on margins, transportation costs, and other transaction costs have been taken from extensive surveys conducted by Huang and Rozelle during the 1990s and the early 2000s. The surveys have also served to identify the commodity price series that provide an appropriate foundation for price comparisons. Some of this China 143 material has been examined in Rozelle, et al. (2000) and Huang, Rozelle, and Chang (2004), who provide information on substantial quality differences between imported and domestic commodities and the resulting biases in price comparisons as a measure of protection. Survey teams from the Center for Chinese Agricultural Policy interviewed traders in 10 cities around China in 2006, and this has provided additional data on more recent years. The data for all the commodities are summarized in appendix B, table B.3. International price data for all commodities except milk--free on board prices and cost, insurance, and freight prices--represent the unit values of exports or imports, with adjustments for quality, where needed. We have taken these data from the Ministry of Commerce and China Customs. For the border price of milk, because no import prices for milk are available, we adjust the farmgate price of milk in New Zealand according to international transportation and insurance rates to create a series for the international price of milk (cost, insurance, and freight included), which we refer to as the reference price. Other data used in this study include tariff rates and information on taxes and subsidies. The tariff rates have been compiled from the Office of Tariff Regulation (1998, 2005). Agricultural tax data have been collected from cost of production surveys conducted by the National Development and Reform Commission. Infor- mation on subsidies (for example, recent grain subsidies) is taken from various government documents. Aggregate input subsidies are based on producer support estimates by the Organisation for Economic Co-operation and Development (OECD), which are disaggregated into individual commodities based on shares in total crop area (see OECD 2005 and the OECD PSE-CSE Database 1986­2005). Results An indirect indication of the success of China's reform policies may be gained by examining the domestic free-market prices for four of the country's most impor- tant crops (Huang et al. 2007). Prices were relatively stable despite the sharp rise in demand after 1980 for all commodities because of increases in the population, rapidly rising incomes, the shift of the population from rural to urban areas, and gradual marketization. The overall self-sufficiency rate for rice, wheat, maize, and soybeans during the 1980s and 1990s was nearly 100 percent (Huang and Chen 1999a, 1999b). It is clear that domestic production was keeping up with the growth in demand. Many studies have found that the rise in supply, as well as in total factor productivity, was generated by a reduction in the distortions in pro- duction incentives that was driven by reforms such as improved property rights and more open access to new technologies (Lin 1992; Fan 1991; Huang and Rozelle 1996; Jin et al. 2002). 144 Distortions to Agricultural Incentives in Asia The role of domestic price and marketing policy It is useful to examine the relationship between the available domestic price series for farmgate procurement and the urban and rural retail prices for the major grain crops (Huang et al. 2007). The importance and role of the government's domestic price and marketing policy for rice, wheat, and maize--the three largest crops in China--may be understood through a comparison of the urban retail price, the rural farmgate procurement price, and the rural retail price. The first two are set by the government, while the third is a free-market price. Until 1992, the urban retail price for rice was usually substantially below the price on the free market in rural areas, despite the costs associated with transferring rice to urban areas. This was a consequence of a procurement price system designed to provide urban residents with relatively inexpensive (that is, subsidized) food. Only urban residents were allowed to buy rice at these low prices, and they were allowed to do so only on the basis of ration coupons that were available in limited quantities. The marketing and procurement system may have been the source of addi- tional distortions. In the 1980s, the relatively low selling price for grain by farmers at the farmgate shows that the food system was transferring income from rural to urban areas. The prices paid to farmers for the output they delivered based on government quotas were far below free-market prices. Nonetheless, given the inframarginal nature of many of the transfers, there is some uncertainty about the effect on the incentives for production and consumption (Sicular 1988b). This is because, after the mid-1980s, farmers were able to sell surplus output at higher market prices once they had met their quota obligations. If a farmer could sell sur- plus grain at an above-quota price determined by market forces, then the farmer may have faced a less severe distortion. However, the inframarginal transfers were not fully decoupled from incentives. It appears, for instance, that they represented an incentive to move out of agriculture (Wang, Rozelle, and Huang 1999). The domestic marketing and procurement system may thus have also distorted incen- tives relative to international prices. Beginning in 1992, however, changes in the domestic marketing and procure- ment system appear to have eliminated this additional layer of regulation for pro- ducers of rice, wheat, and maize. In the early 1990s, the urban price began to rise above the farmgate price, and urban and rural retail prices also moved closer together. This reflects the phasing out of the implicit taxation of farmers through the grain procurement system. The gap between urban and rural retail prices eventually disappeared, and the gap between the rural retail price and the farm- gate price declined, suggesting that there had been an improvement in marketing efficiency (Park et al. 2002). The distortions remaining after the mid-1990s, fol- lowing the disappearance of the distortions caused by the marketing and procure- ment system, reflect only trade policies. China 145 NRAs for the main agricultural commodities In this subsection, we focus on the distortions faced by farmers in China between 1981 and 2005. We first chart the NRAs on output (at the wholesale and farm levels) for each of the 11 commodities. Because input subsidies were generally small during most of our sample period, we do not discuss these measures in this subsection. Each NRA is computed at adjusted exchange rates (see above). The resulting NRAs are summarized in table 3.4 as five-year averages, and annual prices and NRAs are provided in numerous additional tables and figures in Huang et al. (2007). Table 3.4. NRAs for Covered Agricultural Products, China, 1981­2005 (percent) Indicator, product 1981­84 1985­89 1990­94 1995­99 2000­05 Exportablesa,b 58.1 46.3 22.0 0.8 0.2 Rice 55.7 34.0 30.4 6.6 7.2 Fruit 28.5 9.4 4.0 0.0 0.0 Vegetables 41.9 57.5 22.3 0.0 0.0 Poultry 25.1 27.1 2.7 0.0 0.0 Pig meat 78.6 48.8 14.9 0.0 0.0 Import-competing productsa,b 12.0 19.1 1.6 19.3 9.8 Wheat 1.9 22.3 11.3 30.2 4.0 Soybeans 0.6 1.3 4.7 29.5 16.3 Sugar 43.7 44.7 11.7 26.6 29.4 Milk 128.7 58.3 4.3 18.3 24.8 Mixed trade statusa Maize 35.2 16.1 25.1 5.3 12.6 Cotton 33.7 34.6 26.2 3.6 0.7 All covered productsa 50.8 40.6 18.9 2.3 0.9 Distortions Domestic input price 0.3 0.3 0.2 0.7 0.5 distortions Domestic output price 12.6 6.3 6.2 1.1 1.4 distortions Border distortions 38.5 34.6 12.9 2.7 1.8 Dispersion, covered productsc 74.3 52.3 20.7 18.4 15.5 Coverage, at undistorted prices 85 89 85 80 66 Sources: Huang et al. 2007; data compiled by the authors. a. Weighted averages; the weights are based on the unassisted value of production. b. Depending upon their trade status during a particular year, products of mixed trade status are included among exportable products or import-competing products. c. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of covered products. 146 Distortions to Agricultural Incentives in Asia Distortions in the grain economy before 1995. The distortions in the rice economy of China in the 1980s and early 1990s were characterized by two important fea- tures. First, the NRA for rice, an exportable commodity, was negative every year between 1981 and 1995. Ranging between around 60 and 10 percent, the neg- ative NRA, together with the fact that China was able to export rice, suggests that the country was highly competitive in international rice markets during these years. Nonetheless, trade policy kept exporters from shipping large quantities of rice onto world markets and kept the free-market price of rice in China's port cities below the world price. This shows the government's commitment to keeping domestic market prices low. Even if there had been no other distortions in the rice economy, producers would have faced prices below world market prices. The second feature demonstrates that domestic marketing and procurement represented an additional tax on farmers and would have insulated the domestic price of rice from the world market price even if there had been no trade restric- tion at the border. Because of this marketing policy, which lasted to the mid- 1990s, the artificially low government procurement price kept the average price received for rice by farmers systematically below the free-market price. Together, these two features meant that the producer tax on rice ranged between 66 in 1981 and 33 in 1991, making rice producers among the most heavily taxed farmers in China (Huang et al. 2007). Our NRA estimates show that, unlike the case of rice, trade policy offered some protection to wheat between 1981 and the mid-1990s (table 3.4). During this period, the free-market price of wheat in China's port cities ranged between 27 and 64 per- cent higher than the international price of wheat and averaged about 45 percent higher. This alone suggests that wheat producers in China--who are known to pro- duce at a higher cost than producers in many other countries (Huang and Ma 2000)--received significant protection through trade policy. However, domestic marketing policies were having an impact in the opposite direction. The NRA results show how the wheat quotas reduced the potential benefits to farmers from the high rates of protection at the country's borders. Although protection averaged 15 percent for wheat farmers in most years between 1981 and 1995, the rates were zero or even slightly negative in 5 of the 15 years during this period (1981­82, 1990, and 1992­93). The case of maize is a mix of the case of rice and the case of wheat (table 3.4). In some years, trade policy provided positive protection for maize (1983­87, 1989, and 1994), while, in other years, the domestic price of maize was lower than the world market price. On average, trade policy protected maize only marginally between 1981 and 1995. As with rice and wheat, procurement policy depressed the price of maize among farmers. Indeed, except for 1985 and 1994, the net effect of international trade and domestic marketing policy from the 1980s to the early 1990s was to tax maize producers. China 147 Distortions to the grain economy after 1995. Our analysis shows that the govern- ment's international trade and domestic marketing policies changed strikingly after 1995 (table 3.4). The procurement policies that had been taxing rice, wheat, and maize farmers either by adding to the tax imposed by export policy, as in the case of rice, or by reducing protection, as in the case of wheat, were finally elimi- nated from domestic marketing policy. The elimination of the procurement quota contributed significantly to a reduction in the tax burden shouldered by farmers (Huang, Rozelle, and Wang 2006). The liberalization of domestic markets in the mid-1990s was accompanied by a liberalization in trade policy at least in the case of the country's major food grains. After 1995, the taxation and subsidization of rice and wheat were clearly being phased out: the NRAs for rice were rising steadily (they were becoming less negative), and the NRAs for wheat were falling. Probably partly in preparation for the country's accession to the WTO, the government liberalized trade in the main food grains so extensively that, between 1995 and 2001, most of the protection for these crops was phased out. In 2001­05, the NRAs for both rice and wheat were almost zero. The case of maize is a bit different from the situation among other crops. The NRAs for maize have been positive in some of the years since 2000. This protec- tion for maize producers may be partly the result of lobbying by Jilin Province in seeking national support for its most important crop (Rozelle and Huang 2004). Edible oils and cotton. The biggest difference in the analysis of the distortions affecting grain staples and the analysis of the distortions affecting cash crops (at least for soybeans and cotton) is the smaller historical role played by domestic marketing policy among the cash crops. Although procurement delivery quotas aimed at soybean producers existed in some counties, they were not as widespread as the quotas for grains; in many counties, soybeans were not a target of the gov- ernment procurement system. In addition, the implicit tax on soybeans in places where the soybean quotas were collected was lower than the implicit tax on staple grain crops. There is thus little difference between the NRAs on output at the wholesale and farm levels. The same is true of cotton, although the free-market procurement of cotton by private traders was not allowed until the mid-1990s. When reform finally came to the cotton industry, policy makers chose not to adopt a two-tier pricing system. Instead, they opted for both private trade and pri- vate procurement through commercialized government cotton procurement sta- tions. As a result, the NRA measures of distortion on cotton output are nearly the same at the wholesale and farm levels. This is true of all other commodities as well (livestock, horticultural commodities, milk, and sugar). For this reason, the dis- cussion in the rest of this subsection--on the 1980s, the 1990s, and the post-2000 period--focuses only on trade policy. 148 Distortions to Agricultural Incentives in Asia Our analysis shows that, before 1995, soybeans were sometimes taxed and sometimes protected (table 3.4). Although the average level of protection was close to zero, it ranged from 20 percent to 40 percent. Rozelle and Huang (2005) show that much of this fluctuation was caused by domestic production policy, which encouraged soybean production before discouraging it, and then encouraging it again, although the trade regime permitted little trade. However, trends in the NRAs indicate that there has been some movement toward trade lib- eralization in soybeans. Beginning in the late 1990s and continuing through to 2005, the protection rate for soybeans fell from around 30 percent to less than half that amount. This should not be a surprise given the gradual integration of China into world soybean markets and the monotonic rise in imports (which exceeded 25 million tons in 2005). The situation in soybeans stands in sharp contrast to the case of maize, which enjoyed increasing protection. Maize and soybeans compete for land and other resources; it appears that, when the level of protection began to rise for maize, soybean production was significantly liberalized. The NRAs suggest that cotton and rice share similarities. The combination of trade and monopoly procurement policy kept domestic cotton prices lower than world market prices in the 1980s and early 1990s. It appears that planners were taxing cotton farmers to supply the country's emerging textile industry with rela- tively inexpensive raw materials. The high implicit tax on cotton, along with rav- ages caused by insects, undoubtedly contributed to the stagnant and sometimes falling share of cotton in the total planted area in many regions (data of the National Bureau of Statistics). Since 1995, however, (mostly) because of the liberalization in domestic mar- kets and (somewhat) because of trade liberalization, there has been a clear shift in the level of the distortions faced by cotton producers. Although there were fluctu- ations (protection was high in 2000, but cotton was implicitly taxed in 1999 and 2001), the average NRA since the mid-1990s has been close to zero (if, in choosing the border price, one takes into account the trade status during each year). Recently, in most years, despite their authority to impose tariff rate quotas on cot- ton after a threshold amount has been imported, trade officials have essentially allowed the level of imports to be determined by the market. Livestock, sugar, and horticultural commodities. With the exception of fruit for several years in the late 1980s and early 1990s and of poultry before 1985, the pat- terns of distortion in the livestock and horticultural subsectors are remarkably similar (table 3.4). In all cases during the early reform period, there was heavy implicit taxation on livestock and horticultural commodities. As noted by Huang, Rozelle, and Chang (2004), this situation was created partly by the government's grain-first policy. Although China can be competitive in livestock and horticul- tural products, producers were not encouraged to produce or export these China 149 commodities on a large scale. Another part of the problem was the country's own barriers, such as the quotas on exports to Hong Kong, China. Since the late 1990s, the gaps between the domestic and the world prices received by livestock and horticultural producers have narrowed. NRAs have risen toward zero for all these commodities, including pig meat, poultry, vegetables, and fruit. The relaxation in grain-first policies (often called structural adjustment policies) has allowed producers to expand livestock and horticultural production significantly. The main reason for the expansion is the rising demand inside the country (Rosen, Huang, and Rozelle 2004). Nonetheless, China's accession to the WTO and the emergence of a large export-oriented livestock and horticultural segment in the sector have increased the interest and feasibility of participation in international markets. The situation for milk and sugar contrasts with the situation for livestock and horticultural commodities. During the 1980s, the NRAs for milk and sugar were positive and large (table 3.4). The NRAs for milk ranged from 40 to 160 percent between 1980 and 1987. They fell through the early 1990s, but rose again in the late 1990s before falling back somewhat thereafter. The aggregate picture. After aggregating the 11 commodities in our study, we notice a striking pattern in our results (table 3.4 and figure 3.3). In the 1980s and Figure 3.3. NRAs for Exportable, Import-Competing, and All Agricultural Products, China, 1981­2004 35 15 0 5 % NRA, 25 45 65 198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004 year import-competing products total exportables Source: Huang et al. 2007. Note: A total NRA may be above or below the exportable and import-competing averages because the assistance to nontradables and non-product-specific assistance are also included. 150 Distortions to Agricultural Incentives in Asia through to the mid-1990s, import-competing products such as wheat, soybeans, milk, and sugar were only slightly taxed, on average, at the farm level. This conclu- sion takes into account the effect of input subsidies and domestic output price distortions, as well as distortions at the border, the average contributions of which are shown towards the bottom of table 3.4. The rate of protection varied consider- ably across the 11 commodities. It also varied inversely with international price movements, suggesting that the government was seeking to insulate the domestic market a little from world price fluctuations. Exported commodities such as rice, livestock, and horticultural products were subject to substantial rates of taxation, averaging 37 percent in 1981­95, though the variations from year to year were large and included nominal rates of taxation of more than 50 percent in several years. Since exported agricultural products accounted for a greater part of the economy than import-competing agricultural products during the early reform period, the agricultural distortions were negative, on average. In other words, for most of the reform period, the government was taxing farmers through interna- tional trade and domestic marketing policies. One of the main findings of our study is evident in figure 3.3. After 1995, the NRAs of import-competing products fell from around 20 percent to less than 10 percent. During this period, the average NRA for exportables rose (or the implicit taxes on them fell) from about 40 percent to around 15 percent. Taken together, the distortions in agriculture fell to less than 10 percent. In many years, the overall protection was between 0 and 5 percent. Clearly, the combination of domestic marketing reforms and international trade liberaliza- tion generated an agricultural economy that, on average, was neither taxed nor assisted. This does not mean, however, that all distortions were eliminated. In 2000­05, there was still wide dispersion in the NRAs across commodities, several of which continued to show relatively high rates of protection (figure 3.4). For example, the NRAs for sugar and milk were above 20 percent, and the NRAs for soybeans and maize were nontrivial, at more than 10 percent, while the NRAs of the fruit, vegetable, pig meat, and poultry exportable categories were close to zero. The dispersion is much less than in earlier decades, however. This is indicated in the decline in the standard deviation in the NRAs across our 11 commodities, but also in the (anti)trade bias index (tables 3.4 and 3.5). The elimination of distortions has affected food consumers as well. In the 1980s and early 1990s, consumers were being taxed because of the positive protec- tion on import-competing farm products, but, at the same time, they were gaining from being able to consume exportables that were implicitly subsidized by mar- keting and export-restricting policies. Some consumers of rice, wheat, and maize were also receiving additional consumer subsidies. The net effect of these inter- ventions is that, until the late 1990s, consumers were enjoying a large, implicit China 151 Figure 3.4. Average NRAs for Producers of Major Commodities, China, 2000­05 35 30 25 20 % 15 10 NRA, 5 0 5 10 y sugar milk maize pork wheat fruits rice cotton poultr soybeans vegetables major commodity Source: Huang et al. 2007. subsidy, which was reflected in the negative average CTEs in the 1980s (around 40 percent) and early 1990s (around 15 percent). After the mid-1990s, as the distortions against producers declined and markets became the main mechanism for food flows, those households that were net food buyers saw their implicit sub- sidy fall to close to zero overall even though the average CTE for import-competing products was still well above the average for exportables (similar to the NRA: compare tables 3.4 and 3.6). Comparisons with the results of other studies Three recent studies provide information on protection rates for agricultural products in China that may be compared with the estimates in our study. Two of the studies have been conducted by the OECD (2005, 2007); the other study was carried out by the International Food Policy Research Institute (Orden et al. 2007). These studies use a similar methodology, although they differ from our study in covering a much shorter time period, in the specific series used for price comparisons, and in the details of the methodology, including the treatment of exchange rates. The OECD studies cover 1993­2005, while the International Food Policy Research Institute study provides estimates for most commodities from 1995 to 2001 (compare with our coverage, which begins in 1981). 152 Distortions to Agricultural Incentives in Asia Table 3.5. NRAs in Agriculture Relative to Nonagricultural Industries, China, 1981­2005 (percent) Indicator 1981­84 1985­89 1990­94 1995­99 2000­05 Covered productsa 50.8 40.6 18.9 2.3 0.9 Noncovered productsb 29.1 15.4 7.3 7.8 4.2 All agricultural productsa 47.6 37.9 17.2 3.5 2.0 Non-product-specific assistance 2.4 2.4 2.9 3.1 4.0 Total agricultural NRAc 45.2 35.5 14.3 6.6 6.0 Trade bias indexd 0.50 0.55 0.23 0.15 0.07 NRA, all agricultural tradablese 45.2 35.5 14.3 6.6 6.0 NRA, all nonagricultural tradables 41.6 28.3 24.9 9.9 4.7 RRAf 60.6 49.9 31.1 3.0 1.3 Memorandum items: ignoring exchange rate distortionsg NRA, all agricultural products 34.9 27.1 11.6 3.5 2.0 Trade bias index 0.33 0.38 0.13 0.15 0.07 RRA 52.2 41.0 26.5 3.0 1.3 Sources: Huang et al. 2007; data compiled by the authors. a. Including product-specific input subsidies. b. Noncovered import-competing products are assumed to be protected at 75 percent of the rate applying to covered products. Noncovered exportables are assumed to be protected or taxed at 80 percent of the rate applying to covered products. c. Including product-specific input subsidies and non-product-specific assistance. The total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. d. The trade bias index is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. e. Assuming all agricultural production is tradable and including product- and non-product-specific subsidies. f. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. g. The average values of the indicators, as captured by our methodology, if we ignore the distortions in the foreign exchange market, (see Anderson et al. 2008 and appendix A). Broadly comparable estimates for six commodities--rice, wheat, maize, sugar, soybeans, and cotton--in 1995­2001 may be obtained from the three studies.3 In addition, it is possible to compare the OECD and World Bank results for milk over this period. The estimated average protection rates for 1995­2001 are shown in figure 3.5. It is evident from the figure that most of the estimates are similar, and the estimates are close for rice and maize, two extremely important commodities. The estimates for wheat are noticeably different. Our study suggests that wheat was protected, while the other studies suggest it was taxed. Our estimates for China 153 Table 3.6. CTEs for Covered Agricultural Products, China, 1981­2005 (percent) Product 1981­84 1985­89 1990­94 1995­99 2000­05 Exportablesa,b 52.0 47.7 19.2 0.6 0.1 Rice 35.7 28.8 22.3 4.2 0.1 Fruit 29.8 11.8 6.3 0.0 0.0 Vegetables 42.4 57.9 22.5 0.0 0.0 Poultry 26.7 26.1 1.9 0.0 0.0 Pig meat 75.1 46.8 14.9 0.0 0.0 Import-competing productsa,b 16.5 43.6 32.9 25.7 7.4 Wheat 27.5 49.1 43.3 29.0 1.2 Soybeans 4.5 10.3 10.2 26.3 19.2 Sugar 89.1 36.2 32.6 43.0 60.4 Milk 127.2 57.5 4.8 17.7 22.3 Mixed trade statusa Maize 13.2 10.9 13.4 8.4 14.9 Cotton 36.4 37.5 29.0 8.6 6.6 All covered productsa 44.5 41.7 15.8 1.7 0.7 Source: Huang et al. 2007. a. Weighted averages; weights are based on the undistorted value of consumption. b. Depending upon their trade status during a particular year, products of mixed trade status are included among exportable products or import-competing products. sugar, soybeans, and cotton are broadly comparable with the results of the Inter- national Food Policy Research Institute study, but quite different from the OECD estimates. Finally, our estimate of protection for the dairy industry is much lower than the OECD estimate. Careful examination reveals that the differences between our estimates for wheat and the OECD estimates have arisen because of separate assumptions about the quality differentials in domestic and imported wheat. While our study uses a quality adjustment coefficient of 40 percent for domestic wheat relative to imported wheat that is based on Huang, Rozelle, and Chang (2004), the OECD study uses an adjustment coefficient of 15 percent. The much lower OECD esti- mate of the protection for soybeans over this period appears to result primarily from the OECD's use of a Heilongjiang price (a carefully considered decision based on the fact that producers in Heilongjiang Province produce commercially and compete with imports) rather than a national price for soybeans. We find our higher rate of protection in 1993­2001 more plausible given the rapid expansion in imports when the protection subsequently fell. However, the two sets of esti- mates are reassuringly comparable for 2003­05. 154 Distortions to Agricultural Incentives in Asia Figure 3.5. NRA Estimates in Other Studies, by Product, China, 1995­2001 120 % 100 rates, 80 60 annual the 40 of 20 average 0 20 rice wheat maize soybeans milk cotton sugar commodity World Bank OECD IFPRI Source: Huang et al. 2007. Note: IFPRI International Food Policy Research Institute. Our analysis of the distortions on milk uses the New Zealand price for milk as the reference price, plus a transport margin, while the OECD uses measures based on the international trade in processed milk products considered on a milk-equiv- alent basis. Our reference price is notably higher over the period covered, resulting in our lower estimate of the rate of protection. The sizable difference in the two estimates for sugar arises largely because of differences between the farm and wholesale levels in the assumed marketing and handling costs for sugar. Whereas we have assumed that these costs were around 15 percent of the total, the OECD has assumed costs at around 60 percent of the farmgate price. These comparisons of estimates for particular products over a specific period highlight the differences, without emphasizing the important similarities in most of the estimated values, particularly the similarities in the trends over time. How- ever, they do emphasize the importance of examining and reexamining the effect of particular assumptions on NRA estimates. Agricultural Versus Nonagricultural Protection The rate of assistance to a particular sector is an incomplete measure of the impli- cations of the trade regime for outcomes in that sector. The protection in other tradable sectors competing for the same mobile resources such as labor and China 155 capital also matters. In a simple model with only two traded-goods sectors, agri- culture and nonagriculture, a broad idea of the total effect of the prevailing dis- tortions may be obtained using information only on the magnitude of two NRA measures. The distortion in the price of agricultural goods relative to nonagricul- tural goods provides the signal needed to guide the transfer of resources between the two sets of trading activities. For any given level of world prices, this price sig- nal is given by the ratio: (1 NRAagt) (1 NRAnonagt). (3.1) This may be converted into an RRA, as follows: RRA 100*[(100 NRAagt) (100 NRAnonagt) 1]. (3.2) The estimates of NRAnonagt are provided in Huang et al. (2007). These esti- mates take into account the effects of the trade planning mechanism, tariffs, the exchange rate overvaluation prior to 1994, and nontariff measures such as licens- ing and quotas. While extremely simple, we believe that these estimates supply a realistic indication of the broad pattern of incentives in nonagricultural tradables for our sample period. The average rates of assistance to agriculture and to nonagriculture are presented in table 3.5 and figure 3.6. The data show that the agricultural sector was subject to Figure 3.6. NRAs for Agricultural and Nonagricultural Tradables and the RRA, China, 1981­2004 60 40 20 0 cent per 20 40 60 80 198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004 year NRA, agricultural tradables RRA NRA, nonagricultural tradables Source: Huang et al. 2007. Note: For the definition of the RRA, see the text and table 3.5, note f. 156 Distortions to Agricultural Incentives in Asia two strong and reinforcing sets of tax measures in the early years of our sample, one direct and the other indirect. The direct taxation through border measures and the procurement system resulted in a negative NRA of 45 percent in 1981­84. The neg- ative NRA was only slightly lower, at 36 percent, in 1985­89. The indirect taxation of this sector, through support for nonagricultural activities, was similar in magnitude (42 and 28 percent) to the direct taxation during the two periods. The first major change in the incentive environment in the agricultural sector appears to have occurred in the mid-1980s when the rate of protection for nona- gricultural sectors declined. A sharp reduction in the taxation of manufacturing exportables resulted from the widespread availability of duty exemptions for the imported goods used in the production of exports. Protection in the nonagricul- tural sectors appears to have risen slightly during several years in the late 1980s. Increases in tariff rates and in the exchange rate distortions outweighed the reduc- tions in the effects of trade planning and nontariff barriers. However, the protec- tion for nonagricultural sectors declined steadily beginning in 1990. The decline in the direct taxation of the agricultural sector began only later; a sharp fall in the early 1990s occurred, in large part, because of a reduction in the domestic taxation of the agricultural sector via the procurement pricing system (see above). The declines in both the negative protection for agriculture and the positive protection for nonagricultural tradables since the 1990s have dramatically changed the distortions in agricultural incentives. Instead of facing an RRA of close to 50 percent as in the 1980s, farmers are now receiving slightly positive assistance, with an average RRA of 1 percent in 2000­05. Clearly, this phasing out of farmer disincentives has been a major achievement in the government's policy of reform and opening up. Conclusions and Implications The main finding of this study is that the nature and extent of government policy intervention in agriculture in China have changed dramatically over the past 25 years. An agricultural sector characterized by significant distortions has been transformed into a relatively liberal sector. Distortions in external and domestic policies during the early reform period--the 1980s and early 1990s--isolated domestic producers and consumers from international markets. The prices that domestic farmers and consumers faced were nearly independent of trade policy. Domestic pricing and marketing policies forced farmers to sell much of their sur- plus to the government at an artificially lower price even in the case of exportable commodities such as rice (the free-market price of which was close to the interna- tional price). Hence, although there was little trade taxation at the border, domes- tic policies levied a tax on farmers. Similar dynamics characterized the situation China 157 among importable commodities such as wheat and soybeans. Despite fairly high rates of protection through trade policies, producers were receiving much less protection than they would have received had there been a free domestic market for the importables. This occurred even though consumers were implicitly taxed. In contrast, after the late 1980s and early 1990s, the liberalization of domestic markets reduced the distortions deriving from domestic policies. The market has gradually replaced the state as the primary mechanism for allocating resources and has become the basis for the production and marketing decisions of farmers. Especially in the case of importable commodities, trade policy has also become more liberalized, and the distortions arising from border measures have fallen substantially. Thus, we find that China's agriculture has become much less distorted in recent years in two ways. First, the differences between international and domestic mar- ket prices have narrowed considerably for many commodities because of trade policy liberalization. Second, the elimination of domestic policy distortions means that, as trade liberalization allows more imports or exports of agricultural commodities, prices in the domestic market change, and farmers are therefore more directly affected. There are still plenty of distortions to agricultural incentives in China. In some cases, the remaining distortions have arisen from tariffs on importable commodities (for example, wheat and soybeans). In the case of other importable commodities (for example, maize), the use of implicit export subsidies (permitted under WTO rules as domestic marketing, transport, and storage subsidies) represents a wedge between the domestic price and the international price. Thus, some dispersion of NRAs among farm products remained in 2005, but much less than in the 1980s. Notes 1. These findings are based on intensive interviews and surveys conducted by the Center for Chinese Agricultural Policy among a wide range of agricultural ministry personnel, research adminis- trators, research staff, and others involved in the agricultural research system. 2. Raw data on the official exchange rate and the several measures of the secondary-market rate are presented in Huang et al. (2007), together with estimated foreign exchange retention rates, calculated measures of the exchange rates applying, at the margin, to exporters and importers during the period, and estimated equilibrium rates. 3. We thank Andrzej Kwiecinski of the OECD for thoughtful comments that helped us understand the differences between the sets of estimates. References Anderson, K., J. Huang, and E. Ianchovichina. 2004. "Will China's WTO Accession Worsen Farm Household Incomes?" China Economic Review 15 (4): 443­56. 158 Distortions to Agricultural Incentives in Asia Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Ashton, B., K. Hill, A. Piazza, and R. Zeitz. 1984. "Famine in China, 1958­1961." Population and Devel- opment Review 10 (4): 613­45. Center for Price Monitoring. various years. "National Major Commodity (Service) Price Monitoring Data Compilation." Unpublished data compilation, Center for Price Monitoring, National Devel- opment and Reform Commission, Beijing. Chang, G., and G. Wen. 1997. "Communal Dining and the Chinese Famine of 1958­1961." Economic Development and Cultural Change 46 (1): 1­34. China Customs. various years. "Import and Export Statistics." Statistics compilations, Customs Administration, Beijing. deBrauw, A., J. Huang, and S. Rozelle. 2004. "The Sequencing of Reforms in China's Agricultural Transition." Economics of Transition 12 (3): 427­66. deBrauw, A., J. Huang, S. Rozelle, L. Zhang, and Y. Zhang. 2002. "The Evolution of China's Rural Labor Markets during the Reforms." Journal of Comparative Economics 30 (2): 329­53. de Gorter, H., D. Just, and J. Kropp. 2008. "Cross-Subsidization Due to Infra-marginal Support in Agriculture: A General Theory and Empirical Evidence." American Journal of Agricultural Economics 90 (1): 42­54. Dong, X., and G. Dow. 1993. "Monitoring Costs in Chinese Agricultural Teams." Journal of Political Economy 101 (3): 539­53. Fan, S. 1991. "Effects of Technological Change and Institutional Reform on Production Growth in Chinese Agriculture." American Journal of Agricultural Economics 73 (2): 266­75. ------. 1997. "Production and Productivity Growth in Chinese Agriculture: New Measurement and Evidence." Food Policy 22 (3): 213­28. Fan, S., and P. Pardey. 1997. "Research Productivity and Output Growth in Chinese Agriculture." Journal of Development Economics 53 (1): 115­37. Fisman, R., and S. Wei. 2004. "Tax Rates and Tax Evasion: Evidence from `Missing Imports' in China." Journal of Political Economy 112 (2): 471­96. Gale, F. 2006. "Trends in Chinese Food Demand and Trade Patterns." Paper presented at the U.S. Department of Agriculture Agricultural Outlook Forum 2006, "Prospering in Rural America," Arlington, VA, February 16­17. Honma, M., and Y. Hayami. 2008. "Distortions to Agricultural Incentives in Japan, Korea, and Taiwan." Agricultural Distortions Working Paper 35, World Bank, Washington, DC. Hu, R., and J. Huang. 2001. "Investment in Agro-Technology Extension and Its Effect in China." Strategy and Management 3: 25­31. Huang, J., and H. Bouis. 1996. "Structural Changes in Demand for Food in Asia." Food, Agriculture, and the Environment Discussion Paper 11, International Food Policy Research Institute, Washington, DC. Huang, J., and C. Chen. 1999a. "Effects of Trade Liberalization on Agriculture in China: Institutional and Structural Aspects." CGPRT Centre Working Paper 42, Regional Co-ordination Center for Research and Development of Coarse Grains, Pulses, Roots, and Tuber Crops in the Humid Tropics of Asia and the Pacific, United Nations Economic and Social Commission for Asia and the Pacific, Bogor, Indonesia. ------. 1999b. "Effects of Trade Liberalization on Agriculture in China: Commodity Aspects." CGPRT Centre Working Paper 43, Regional Co-ordination Center for Research and Develop- ment of Coarse Grains, Pulses, Roots, and Tuber Crops in the Humid Tropics of Asia and the Pacific, United Nations Economic and Social Commission for Asia and the Pacific, Bogor, Indonesia. Huang, J., and H. Ma. 2000. "International Comparison of Agricultural Prices." International Trade 10: 20­24. Huang, J., and S. Rozelle. 1996. "Technological Change: Rediscovering the Engine of Productivity Growth in China's Rural Economy." Journal of Development Economics 49 (2): 337­69. China 159 ------. 2002. "China's Accession to WTO and Likely Shifts in the Agriculture Policy." Working paper, Center for Chinese Agricultural Policy, Chinese Academy of Sciences, Beijing. Huang, J., S. Rozelle, and M. Chang. 2004. "The Nature of Distortions to Agricultural Incentives in China and Implications of WTO Accession." World Bank Economic Review 18 (1): 59­84. Huang, J., S. Rozelle, W. Martin, and Y. Liu. 2007. "Distortions to Agricultural Incentives in China." Agricultural Distortions Working Paper 29, World Bank, Washington, DC. Huang, J., S. Rozelle, and H. Wang. 2006. "Fostering or Stripping Rural China: Modernizing Agriculture and Rural to Urban Capital Flows." Developing Economies 44 (1): 1­26. Jin, S., J. Huang, R. Hu, and S. Rozelle. 2002. "The Creation and Spread of Technology and Total Factor Productivity in China's Agriculture." American Journal of Agricultural Economics 84 (4): 916­39. Kiguel, M., and S. O'Connell. 1995. "Parallel Exchange Rates in Developing Countries." World Bank Research Observer 10 (1): 21­52. Lardy, N. R. 1983. Agriculture in China's Modern Economic Development. Cambridge: Cambridge University Press. ------. 1992. Foreign Trade and Economic Reform in China, 1978­90. Cambridge: Cambridge University Press. ------. 2001. Integrating China into the Global Economy. Washington, DC: Brookings Institution. Lin, J. 1991. "Prohibitions of Factor Market Exchanges and Technological Choice in Chinese Agricul- ture." Journal of Development Studies 27 (4): 1­15. ------. 1992. "Rural Reforms and Agricultural Growth in China." American Economic Review 82 (1): 34­51. Lin, J., F. Cai, and Z. Li. 1996. The China Miracle: Development Strategy and Economic Reform. Hong Kong, China: Chinese University Press. Lin, J., and D. Yang. 2000. "Food Availability, Entitlement and the Chinese Famine of 1959­61." Economic Journal 110 (460): 136­58. Lyons, T. 1987. Economic Integration and Planning in Maoist China. New York: Columbia University Press. Martin, W. 2002. "Implications of Reform and WTO Accession for China's Agricultural Policies." Economics of Transition 9 (3): 717­42. McMillan, J., J. Whalley, and L. Zhu. 1989. "The Impact of China's Economic Reforms on Agricultural Productivity Growth." Journal of Political Economy 97 (4): 781­807. Ministry of Agriculture. various years. China Agricultural Yearbook. Beijing: China Agricultural Press. ------. various years. China Agricultural Development Report. Beijing: China Agricultural Press. Ministry of Commerce. various years. China Foreign Trade Yearbook. Beijing: China Foreign Economic and Trade Press. National Bureau of Statistics of China. various years. China Rural Statistical Yearbook. Beijing: China Statistics Press. ------. various years. China Yearbook of Agricultural Price Survey. Beijing: China Statistics Press. ------. various years. Statistical Yearbook of China. Beijing: China Statistics Press. National Development and Reform Commission. various years. Agricultural Production Cost and Revenue Materials Compilation. Beijing: China Statistics Press. Nickum, J. 1998. "Is China Living on the Water Margin?" China Quarterly 156 (December): 880­98. Nyberg, A., and S. Rozelle. 1999. Accelerating China's Rural Transformation. Washington, DC: World Bank. OECD (Organisation for Economic Co-operation and Development). 2005. OECD Review of Agricultural Policies: China. Paris: OECD. ------. 2007. Agricultural Policies in Non-OECD Countries: Monitoring and Evaluation 2007. Paris: OECD. OECD PSE-CSE Database (Producer and Consumer Support Estimates, OECD Database 1986­2005). Organisation for Economic Co-operation and Development. http://www.oecd.org/document/55/ 0,3343,en_2649_33727_36956855_1_1_1_1,00.html (accessed December 2007). Office of Tariff Regulation of the State Council. 1998. Import and Export Tariff Regulation of the People's Republic of China, 1996, 1997 and 1998. Beijing: Law Press. 160 Distortions to Agricultural Incentives in Asia ------. 2005. Import Tariff and Export Tariff Rebate Compilation 2005. Beijing: Economic and Scientific Press. Orden, D., F. Cheng, H. Nguyen, U. Grote, M. Thomas, K. Mullen, and D. Sun. 2007. Agricultural Producer Support Estimates for Developing Countries: Measurement Issues and Evidence from India, Indonesia, China, and Vietnam. Research Report 152. Washington, DC: International Food Policy Research Institute. Park, A., H. Jin, S. Rozelle, and J. Huang. 2002. "Market Emergence and Transition: Arbitrage, Transi- tion Costs, and Autarky in China's Grain Market." American Journal of Agricultural Economics 84 (1): 67­82. Putterman, L. 1993. Continuity and Change in China's Rural Development. New York: Oxford University Press. Rosen, D., J. Huang, and S. Rozelle. 2004. Roots of Competitiveness: China's Evolving Agriculture Interests. Washington, DC: Institute for International Economics. Rozelle, S., and J. Huang. 2004. "China's Maize Economy: Supply, Demand and Trade." Report, U.S. Grains Council, Beijing. ------. 2005. "China's Soybean Economy: Supply, Demand, and Trade." Report, American Soybean Association, Beijing. Rozelle, S., J. Huang, and D. Sumner. 2006. "Supply, Demand and Trade for Horticultural Commodi- ties in China." Working paper, Department of Agricultural and Resource Economics, University of California Davis, Davis, CA. Rozelle, S., A. Park, J. Huang, and H. Jin. 2000. "Bureaucrat to Entrepreneur: The Changing Role of the State in China's Transitional Commodity Economy." Economic Development and Cultural Change 48 (2): 227­52. Rozelle, S., and J. Swinnen. 2004. "Success and Failure of Reform: Insights from the Transition of Agri- culture." Journal of Economic Literature 42 (2): 404­56. Shan, W. 1989. "Reforms of China's Foreign Trade System: Experiences and Prospects." China Economic Review 1 (1): 33­55. Sicular, T. 1988a. "Agricultural Planning and Pricing in the Post-Mao Period." China Quarterly 116 (December): 671­705. ------. 1988b. "Plan and Market in China's Agricultural Commerce." Journal of Political Economy 96 (2): 283­307. ------. 1995. "Redefining State, Plan, and Market: China's Reforms in Agricultural Commerce." China Quarterly 144 (December): 1020­46. Stone, B. 1988. "Developments in Agricultural Technology." China Quarterly 116 (December): 767­822. Stone, B., and S. Rozelle. 1995. Foodcrop Production Variability in China, 1931­1985. Research Notes and Studies, vol. 9. London: School of Oriental and African Studies, University of London. Timmer, C. P. 1998. "The Agricultural Transformation." In International Agricultural Development, 3rd ed., ed. C. Eicher and J. Staatz, 113­35. Baltimore: Johns Hopkins University Press. UNCTAD (United Nations Conference on Trade and Development). 1994. Monitoring Import Regimes. Part 1 of Directory of Import Regimes. New York: United Nations. Wang, D., S. Rozelle, and J. Huang. 1999. "The Impact of Procurement Quota on Agricultural Production in China's Agriculture." Working paper, Center for Chinese Agricultural Policy, Chinese Academy of Agricultural Sciences, Beijing. Wang, H., X. Dong, J., Huang, S. Rozelle, and T. Reardon. 2006. "Producing and Procuring Horticul- tural Crops with Chinese Characteristics: A Case Study in the Greater Beijing Area." Working paper, Center for Chinese Agricultural Policy, Chinese Academy of Sciences, Beijing. Wang, J., J. Huang, and S. Rozelle. 2005. "Evolution of Tubewell Ownership and Production in the North China Plain." Australian Journal of Agricultural and Resource Economics 49 (2): 177­95. Wang, J., Z. Xu, J. Huang, and S. Rozelle. 2005. "Incentives in Water Management Reform: Assessing the Effect on Water Use, Productivity, and Poverty in the Yellow River Basin." Environment and Devel- opment Economics 10 (6): 769­99. China 161 Watson, A. 1994. "China's Agricultural Reforms: Experiences and Achievements of the Agricultural Sector in the Market Reform Process." Working Paper 94/4, Chinese Economic Research Unit, University of Adelaide, Adelaide, Australia. Wen, G. 1993. "Total Factor Productivity Change in China's Farming Sector: 1952­1989." Economic Development and Cultural Change 42 (1): 1­41. World Bank. 1985. World Development Report 1985: International Capital and Economic Development. New York: Oxford University Press; Washington, DC: World Bank. ------. 1994. China: Foreign Trade Reform. World Bank Country Study. Washington, DC: World Bank. Yao, S. 1994. Agricultural Reforms and Grain Production in China. London: Macmillan. Zhang, L., Q. Li, and S. Rozelle. 2005. "Fertilizer Demand in China: What is Causing Farmers to Use So Much Fertilizer?" Working paper, Center for Chinese Agricultural Policy, Chinese Academy of Sci- ences, Beijing. Part III II SOUTHEAST ASIA 4 INDONESIA George Fane and Peter Warr In Indonesia, agricultural trade policy is a politically charged subject. Rice, the sta- ple food, is a net import, and this one commodity has been a central focus of the government's food policy throughout the postindependence period. Self- sufficiency in rice, meaning the elimination of rice imports, has been a cherished goal of agricultural policy for all this time. It is an emotive subject, closely linked in the public imagination to Indonesian nationalism. When asked to name his proud- est single achievement, Soeharto, the president of Indonesia for 32 years, until 1998, cited the country's self-sufficiency in rice.1 The dominance by rice is evident in this study, which documents the changing structure of agricultural protection in Indonesia and attempts to explain the forces that have driven the changes. Our four central themes may be summarized as follows. First, the variations in distortion by sector have been driven by the government's wish to (a) be self- sufficient in food, (b) stabilize food prices at acceptable levels, and (c) promote manufacturing. (Food processing has been an important component of manufac- turing; it was even more important in the 1970s and 1980s than it is now.) These aims led to taxes on unprocessed exports and to subsidies for processing. Two import-competing industries--sugar and rice--have been significantly protected, and the rates of this protection have increased in recent years. The rate of protec- tion of the sugar industry is particularly high. Growers of rice have also received protection, but, until around 2000, this occurred mainly through input subsidies The authors gratefully acknowledge the excellent research assistance of Arief Ramayandi; the helpful comments and data assistance of Neil McCulloch, Rina Oktaviani, Peter Rosner, and Peter Timmer; the useful comments of workshop participants; and the significant work with spreadsheets undertaken by Marianne Kurzweil and Ernesto Valenzuela. 165 166 Distortions to Agricultural Incentives in Asia rather than through the price for the product. This protection may be explained through our hypotheses both as the main element in food self-sufficiency and as a result of the effect rice prices have on manufacturing wages through the cost of living. Since 2000, the rice industry has become more tightly protected; rice imports have been banned. Second, during the long presidency of Soeharto, the country could afford aims (a) to (c) under good fiscal conditions, but not under bad fiscal conditions. So, good fiscal conditions meant more protection and bad trade policies, while bad fiscal conditions meant less protection and good trade policies. Third, growing evidence from around the world convinced policy makers in Indonesia and most other East Asian countries to rely more on markets and less on government intervention. This evidence was based on theoretical arguments, statistical studies, and simple two-country comparisons such as Myanmar and Thailand, the German Democratic Republic and the Federal Republic of Germany, Austria and Hungary, and the Democratic People's Republic of Korea and the Republic of Korea. The resulting policy shift was influenced by the collapse of the communist system and the victory of capitalism in the Cold War. This may have contributed to the long-term shift worldwide toward less manufacturing protec- tion and less agricultural export taxation beginning around the mid-1980s. Fourth, following the democratic reforms that occurred in the wake of the Asian financial crisis of the late 1990s, agricultural protectionism increased some- what in Indonesia. Aggregate measures of protection indicate that these changes, along with reduced protection in manufacturing, caused a switch in the agricul- tural sector from net taxation to net subsidization (by a narrow margin) relative to manufacturing. However, these aggregate measures mask the fact that agricul- tural protection is concentrated in only two crucial industries: sugar and rice. The next section describes the changing structure of the Indonesian economy, with an emphasis on the agricultural sector. The subsequent two sections provide overviews on government economic policy in general and on government policies toward agriculture during the period since independence. In the following sec- tions, we attempt to supply a political econometric explanation for the changing structure and pattern in agricultural distortions over the last three decades, esti- mates of which we provide for individual farm industries and for agriculture as a whole. The final section concludes. Economic Growth and Structural Change From 1968 to 2005, Indonesia's gross domestic product (GDP) grew in real terms at an average annual rate of 6.3 percent. The broad characteristics of this growth are summarized in table 4.1. For ease of comparison with other Asian economies, Indonesia 167 Table 4.1. Real GDP Growth and Its Sectoral Components, Indonesia, 1968­2005 (annual average percent) Preboom, Boom, Crisis, Recovery, Total, Sector 1968­86 1987­96 1997­99 2000­05 1968­2005 Total GDP 7.4 7.7 2.5 4.6 6.3 Agriculture 4.4 3.4 0.6 3.5 3.7 Industry, including mining 10.6 9.8 2.3 4.2 8.5 Services 7.8 7.9 4.0 5.7 6.6 Source: Author calculations based on World Development Indicators Database 2008. the table distinguishes between the preboom period prior to 1987 and the 10 boom years that preceded the Asian crisis of the late 1990s. During the precrisis decade of 1987­96, many Asian countries witnessed growth that was far more rapid than the growth in preceding decades. Indonesia also grew rapidly during this decade, but, as the table shows, the growth was only marginally more rapid than the growth during the previous two preboom decades. The country's eco- nomic growth had been sustained over several decades. Output contracted during the crisis years, and, during the subsequent recovery period, growth averaged a more moderate 4.6 percent (see Fane and Warr 2007 for additional evidence on points discussed in this section). As typically occurs in rapidly growing economies, agricultural output expanded more slowly than GDP, implying that there was a declining share of agriculture in aggregate output. The agricultural sector accounted for 56 percent of GDP in 1965. By 2004, this share had declined to 15 percent. Over the same period, the GDP share of manufacturing and other industry rose from 13 to 44 percent, and the share of services rose from 31 to 41 percent. For a more detailed study of the changing composition of the agricultural sector, one may conveniently use the BPS input-output tables that are available on Indonesia for 1971, 1980, 1990, and 2000. As incomes rise, the share of spending on starchy sta- ples usually falls, while the share of spending on meat, fruits, and vegetables usu- ally rises. The Indonesian experience fits this common pattern. The output growth within agriculture was achieved despite a rapidly diminish- ing share of the supply of labor and capital. Furthermore, while agriculture grew more slowly than other sectors during boom periods, the growth rate also declined less during the crisis years than the corresponding rates of growth in other sectors. The agricultural sector has acted as a shock absorber; this was par- ticularly important during the crisis, when agricultural employment absorbed the large numbers of people laid off in urban centers. Although GDP grew much 168 Distortions to Agricultural Incentives in Asia Table 4.2. Subsector Shares of Agricultural Value Added, Indonesia, 1971­2000 (percent) Subsector 1971 1980 1990 2000 Paddy 46.1 38.0 37.5 30.8 Maize 3.1 3.7 4.1 5.9 Root crops 7.2 6.8 7.6 8.9 Fruits and vegetables 14.1 14.5 21.7 21.8 Other food crops 3.3 4.4 6.4 3.9 Rubber 5.5 5.2 2.0 5.5 Sugarcane 2.2 2.4 2.1 2.5 Coconut 5.2 4.3 3.3 3.7 Palm oil 2.9 2.1 2.4 2.3 Tobacco 2.5 1.7 0.7 0.3 Coffee 2.6 4.3 1.5 0.9 Tea 1.4 1.9 0.5 0.3 Cloves 1.4 3.0 1.6 0.9 Other agriculture 1.8 1.7 3.5 7.3 Livestock 0.6 6.0 5.0 4.9 Total 100.0 100.0 100.0 100.0 Source: BPS 1971, 1980, 1990, 2000. more slowly during the recovery in 2000­05 than it had during the boom decade, agricultural growth did not fall with respect to the boom years. Table 4.2 summarizes the changing composition of the value added in agricul- ture since 1971 based on data from the input-output tables. The output of paddy (unmilled rice produced at the farm level) contracted from 46 to 31 percent of agricultural value added, while the share of fruits and vegetables increased from 14 to 22 percent and the share of livestock rose from 0.6 to 5 percent. It is somewhat surprising that the shares of intermediate inputs used in agri- culture actually contracted. The apparent reason is that fertilizer and pesticide usage was subsidized from the late 1960s until the late 1980s under a government rice-intensification program called Bimas (discussed below). When the subsidies were phased out, fertilizer and pesticide use contracted markedly, especially in rice production. Most intermediate goods used in agriculture are domestically pro- duced. Between 1980 and 2000, the annual share of imported intermediate goods in total intermediate goods used in the country increased from 3.8 to 10.2 percent. In 1971, sales of paddy by farmers to intermediate users (rice millers) accounted for 56 percent of the total value of paddy output, implying that almost half of the output was being milled by individual households. By 2000, sales to Indonesia 169 Table 4.3. Export Sales in Total Sales and Imports in Total Usage, Indonesia, 1971­2000 (percent) Export share Import share Subsector 1971 1980 1990 2000 1971 1980 1990 2000 Paddya 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Maize 11.4 0.3 1.8 0.3 0.0 1.1 0.2 11.3 Root crops 4.7 3.7 0.7 0.3 0.0 0.1 0.0 0.1 Fruits and vegetables 0.1 0.2 0.2 0.3 0.6 1.1 0.9 4.9 Other food crops 4.3 0.3 1.9 0.4 0.0 22.7 18.1 47.7 Rubber 57.5 58.4 6.5 0.6 0.1 0.0 0.1 0.7 Sugarcane 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Coconut 6.6 1.2 0.3 1.5 0.0 0.0 0.0 0.1 Palm oil 2.3 33.5 29.5 0.2 0.0 0.0 0.1 0.2 Tobacco 7.4 12.3 7.6 0.0 1.6 7.3 15.6 0.0 Coffee 21.5 58.7 13.7 0.0 0.0 0.0 0.0 0.0 Tea 24.5 18.5 32.5 0.0 0.0 0.0 3.9 0.0 Cloves 0.0 0.0 0.1 3.0 47.4 12.7 0.0 24.5 Other agriculture 40.5 15.9 26.7 21.2 24.1 43.8 1.4 30.3 Livestock 1.2 0.0 1.8 2.1 0.0 0.5 0.9 8.4 Total Agriculture 7.5 9.3 2.8 2.6 1.3 2.8 2.0 8.7 Milled ricea 0.0 0.8 0.0 0.0 13.5 12.5 0.2 3.9 Fertilizers and 0.0 4.4 14.6 22.9 66.0 18.1 9.6 23.3 pesticides Source: BPS 1971, 1980, 1990, 2000. a. The input-output tables classify paddy (unmilled rice) as an output of the agricultural sector and milled rice as an output of the manufacturing sector intermediate users accounted for 98 percent of the total value of the output. Sim- ilar trends occurred in maize, rubber, sugarcane, palm oil, coffee, and tea. The international trading position of the major agricultural commodities is summarized in table 4.3. The most important import-competing agricultural products are rice, sugar, maize, and soybeans. Major export-competing products include coffee, rubber, tobacco, tea, palm oil, copra, shrimps, and spices. Paddy is neither exported nor imported, but milled rice has historically been a major import item. Since 2002, rice imports have been officially banned, but some imports have still occurred. Cassava is mainly nontraded, although there are exports of its derivatives, manioc and tapioca. Much of the livestock subsector is also nontraded, although chickens are exported, while beef and dairy products compete with imports. 170 Distortions to Agricultural Incentives in Asia Over time, some sectors have shifted from one trade category to the other (import-competing or exportable). The key example is sugar, which was the most important export during the colonial era, but has become one of the most highly protected import-competing products in the postindependence period. Another example is maize, which switched from net export status to net import status during the 1990s. Fruits and vegetables have become major net imports, as have soybeans (which are included in the government's input-output tables as "other food crops"). Policy Evolution Indonesia obtained its independence from the Netherlands in 1949. The next two decades were chaotic. The postindependence government of President Soekarno pursued a nationalistic, quasi-socialist economic policy that produced hyperinfla- tion and economic stagnation. In 1966, Soekarno was displaced amid economic chaos by one of his generals, Soeharto, whose regime, called the New Order, lasted until the macroeconomic crisis of 1998. Soeharto pursued more market-oriented economic policies than his predecessor had. On assuming power, Soeharto speed- ily introduced a macroeconomic stabilization program and then began liberaliz- ing trade and investment policies. In 1967, foreign investors were guaranteed the right to repatriate both capital and profit, and, beginning in 1970, the capital account was almost completely open. As we see below, trade policy under Soeharto's government was much less open. It was characterized by the taxation of exports, especially nonfood agricultural exports, and the protection of imports, including some food imports. In the wake of the commodity boom of 1972­73 and the oil price shocks of 1973­74 and 1979­80, trade policy became increasingly inward-looking. These external events tripled the ratio of Indonesia's export prices to its import prices. Between the early 1970s and the mid-1980s, the government taxed or banned some traditional exports, pursued self-sufficiency in rice, and used part of the burgeoning oil revenues to establish import-substituting manufacturing indus- tries, which it then protected. In the early to mid-1980s, several traditional export industries were subjected to quantitative trade restrictions. The restrictions included a ban on log exports that conferred high rates of effective protection on the plywood manufacturing industry, for which raw timber is the principal input. Licensing systems were introduced for exports of vegetable oils, several spices, cof- fee, and some grades of rubber. In the case of palm oil, domestic refiners were pro- tected by a tax on exports and a requirement that growers supply these refiners with part of their output at low, controlled prices. In 1982, the price of petroleum began to decline. By the mid-1980s it had fallen from US$28 to US$10 per barrel. Many oil-exporting countries, including Nigeria Indonesia 171 and República Bolivariana de Venezuela, were unable to adjust to these external changes without devastating domestic consequences, but Indonesia responded quickly by cutting public spending and devaluing its currency, partly to promote non-oil exports. In addition, a value added tax was introduced between 1983 and 1986. At first, trade policy became increasingly oriented toward import substitu- tion, and the system of import licensing was extended. After this initial protec- tionist response to lower petroleum export revenues, however, trade policy was significantly liberalized starting in 1985. With the stated goal of promoting non-oil exports, the government introduced a series of reforms that reduced tariffs and nontariff barriers (NTBs). Following tariff cuts in 1985, the government transferred most customs functions from the Indonesian Customs Service to an international inspection company, SGS of Switzerland. The role of SGS was phased out by 1995. NTBs on imports were pro- gressively relaxed beginning in 1986, and the system of providing exporters with duty-free inputs was extended. According to the estimates in table 4.4, the effective rate of protection in agriculture declined from 24 to 12 percent between 1987 and 1995, while, in manufacturing, it declined much farther, from 86 to 24 percent, over the same period. Since there was probably more water (unused protection) in the tariffs in 1987 than in 1995, the true reductions in protection were probably somewhat smaller than these numbers indicate, but the decline was still substantial. Mean- while, many NTBs were replaced by tariffs. The coverage of restrictive NTBs Table 4.4. Estimated Effective Rates of Protection, Indonesia, 1987, 1995, and 2003 (percent) Sector, indicator 1987 1995 2003 Agriculture 24 12 9 Manufacturing, excluding oil and gas 86 24 16 Manufacturing, including oil and gas 48 20 13 All tradable sectors 18 8 4 Antitrade biasa 52 28 20 Source: 1987­95: Fane and Condon 1996; 2003: author estimates (Fane). Note: The estimates for 1987 and 1995 measure the rates shortly before the reform package of December 1987 and shortly after the reform package of May 1995. The estimates for 2003 apply the same methodology, but using the tariff cuts announced in the May 1995 package for implementation by 2003. With relatively minor exceptions, the plans announced in 1995 were implemented. The projections of the rates for agriculture in 2003 make no allowance for the increased protection of rice and sugar that occurred beginning in 2000. a. The antitrade bias (ATB) is defined as: 1 ATB (1 gm) (1 ge), where gm and ge denote the average rate in, respectively, all import-competing sectors and all export-competing sectors. 172 Distortions to Agricultural Incentives in Asia declined from 44 percent of total value added in all traded industries in 1986 to 23 percent in 1995. This switch from NTBs to tariffs was more extensive in man- ufacturing than in agriculture, where the coverage of NTBs declined from 67 to 48 percent. In the wake of the financial crisis of 1997­98, the government was obliged to allow free imports of both rice and sugar as a condition for borrowing from the International Monetary Fund (IMF). However, since the end of the IMF program, imports of rice and sugar have again been restricted by both tariffs and NTBs. Agricultural Protection by Subsector The distinction between import-competing products and export products is not always entirely clear-cut, but it is nevertheless crucial to any discussion of Indone- sian agricultural policies. Whereas import-competing production has generally been protected by government policies, export-competing production has gener- ally been taxed. For centuries, the country's major staple food crops have been net imports. Both the Dutch colonial government and the government of Indonesia after independence generally tried to control the price of rice and other important food crops to balance the competing interests of domestic producers and con- sumers. Except when the world prices of food crops have been unusually high, imports have been directly restricted, or subject to tariffs, or both. On the rare occasions when world prices have been so high that growers would have had an incentive to export food crops such as rice, they have usually been prevented from doing so by nontax measures, including outright export bans. In contrast, export crops have been seen by successive governments over the last two centuries as a useful source of tax revenue. Under the cultivation system (cultuurstelsel) introduced by the Dutch in 1830, the production of cash crops for export was stimulated by imposing taxes on villagers that could be most easily paid in kind by handing over crops that the Dutch East India Company then processed and exported. By far the most important of these exports was sugar; other important exports in the 19th century were coffee, tea, indigo, and cinna- mon. Booth (1988) reports that, in the late 1830s, 40 percent of the total income of the government of the Netherlands was derived from the cultivation system in the Indonesian colonies. Since independence, the Indonesian government's revenue from export crops has been obtained through export taxes that have tended to depress the domestic production and the exports of the relevant crops. The main reason for raising export revenue in this quite different way has presumably been the government's desire, since independence, to promote the development of the manufacturing sector, of which food processing is an important part. Rice, sugar, and soybeans Indonesia 173 have been protected from import competition by NTBs. The remainder of import-competing agriculture has been protected by tariffs and tariff surcharges. (Rice, sugar, and soybeans, along with maize, are discussed in the next four sub- sections. Import-competing and export agriculture are discussed in the subse- quent subsections.) Rice The most important and most enduring NTBs have been those on rice and sugar. Chart a in figure 4.1 shows estimates of the domestic wholesale prices and border prices for rice.2 (All the price series in figures 4.1­4.4 are in rupiah per kilogram, divided by the GDP deflator indexed at 1 in 2005. The nominal rate of protection [NRP] from import competition is the percentage by which the domestic price exceeds the border price.) While there have been enormous nominal increases in rice and sugar prices since the early 1970s, figures 4.1 and 4.2 show that any trends in the real prices of these products have been relatively small. It is clear from chart a in figure 4.1 that the domestic wholesale price of rice has fluctuated much less than the border price and that domestic prices have not dif- fered greatly, on average, from the trend level of border prices. Price stabilization was achieved by giving the government's food logistics agency, Bulog (Badan Urusan Logistik), a monopoly over international trade in rice and directing the agency to build up buffer stocks to smooth out fluctuations in domestic supply. It is significant that this stabilization of domestic prices was achieved while keeping the trend value of domestic prices roughly in line with the trend in world prices. Average rates of protection in the output market were low, but input markets were another matter. Beginning in the 1970s, the Soeharto government used part of its new oil wealth to promote self-sufficiency in rice by subsidizing the adoption of high- yielding varieties that had been made available through the Green Revolution. These new varieties required greatly increased use of irrigation, fertilizers, and pesticides, which the government helped to finance. An important motivation for this policy was the fear of a repetition of the riots precipitated by high food prices in 1965. Under the Bimas (bimbingan masal or massive guidance) program, which was introduced with the explicit goal of rice self-sufficiency, farmers received agricul- tural extension services and subsidized credit, seeds, fertilizers, and pesticides. The government also upgraded irrigation facilities and expanded irrigation coverage. The resulting rise in the profitability of rice growing, together with some coercion among farmers who were reluctant to extend the area of rice cultivation, led to a 17 percent increase in gross harvested area in the decade to 1985.3 The increase in Figure 4.1. Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Rice and Urea Fertilizer, Indonesia, 1975­2005 a. Rice b. Urea fertilizer Source: Fane and Warr 2007. 174 Indonesia 175 cultivated area, together with a 50 percent rise in average yields during the same period, allowed Bulog to reduce domestic rice prices relative to the consumer price index between the late 1970s and 1985, while gradually phasing out imports and then halting imports altogether in 1985. Lower world oil prices and advice from the World Bank contributed to the reduction in agricultural input subsidies in the late 1980s and early 1990s. Graph b in figure 4.1 indicates the fall in the real price of urea from the late 1970s to the early 1980s and the subsequent rise in the domestic wholesale price of urea relative both to the consumer price index and to the border price in the late 1980s and early 1990s. Under an export licensing scheme, exports of urea require special approval from the Ministry of Trade. The year-to-year determination of the mag- nitude of the licensed exports is not transparent, but the ministry tends to place priority on ensuring that domestic supplies are stable at a price lower than the world market price. This results in the negative rates of protection from import competition shown in figure 4.1, graph b. In the late 1980s, the strict policy of zero imports of rice was replaced by a pol- icy involving borrowing rice from Vietnam in times of shortage and repaying the rice loans in times of surplus. These loans were conducted in bilateral govern- ment-to-government deals in which Bulog represented the Indonesian govern- ment. In the early 1990s, it gradually became apparent that Indonesia was unable to maintain rice self-sufficiency even on average over a period of years. To satisfy domestic demand at acceptable prices, Bulog was obliged to undertake substantial net imports. When the Asian crisis forced the government to borrow from the IMF in 1997, one of the loan conditions to which the government agreed was the removal of Bulog's monopoly on rice imports. Until 1999, there was also no import duties on rice, but the IMF's aim of free trade in rice proved illusory because the financial crisis briefly converted rice into a potential export and the government banned exports to reduce pressure on domestic prices. Graph a in figure 4.1 shows that border prices, converted to rupiah at the devalued exchange rate, were far above domestic prices in 1998. This occurred because the massive depreciation of the exchange rate between mid-1997 and mid-1998 initially outweighed the much more gradual rise in domestic prices. This episode clearly demonstrates that the government's policy has always been to stabilize food prices at acceptable levels rather than simply to protect growers. The general increase in domestic prices in 1998­99 and the stabilization of the exchange rate after mid-1998 removed the incentive to export rice. Bulog's monopoly on imports was not immediately reimposed, but a 20 percent tariff on rice imports was introduced in 1999. Problems with underinvoicing by importers resulted in this tariff being converted to a specific tariff at Rp 430 per kilogram. In 176 Distortions to Agricultural Incentives in Asia 2002, Bulog's monopoly over imports was restored, and, since 2004, rice imports have officially been banned, although Bulog has occasionally been issued with special import permits. Sugar The Indonesian sugar industry is dominated by the state-owned mills, mainly on Java, that were acquired through the nationalization of the formerly Dutch- owned sugar estates in 1957. Investment and technical progress in this sector have been extremely sluggish, and the industry has languished behind protective barri- ers. The finished product of the antiquated factories--known as mill white or plantation white--is not precisely comparable to the refined sugar or the raw sugar traded on the world market. Plantation white contains more impurities, mainly molasses, than internationally traded raw sugar, but has already under- gone some of the bleaching processes that separate refined sugar from raw sugar in more technologically advanced sugar industries. Most firms in the food and beverage subsectors are unable to use plantation white because of its relatively high level of impurities; their needs are mainly met by imports of raw sugar, although a small amount of raw sugar is produced domestically. As in the case of rice, the main motive behind government policy on sugar appears to be the desire to stabilize the domestic price at an acceptable level. In addition, in the case of sugar, the government has tried to protect the sugar facto- ries that it owns. This may explain why, at least since 1957, the sugar industry has been more tightly regulated than any other agricultural sector: the government has monopolized not only imports, but also domestic marketing. Government ownership likewise helps to explain newspaper reports that, in the 1970s, farmers in traditional sugar-growing areas were regularly forced to grow sugar to supply local factories because of threats that their other crops would be burned. Figure 4.2 compares the border price of raw sugar (after allowing for margins between the free on board price and the domestic wholesale price) with the domestic wholesale price of plantation white. The figure shows that, for much of the period since 1970, the domestic price was about twice the border price, imply- ing an NRP of about 100 percent. However, in 2006, this gap was narrowed appre- ciably because of the abrupt rise in world prices. Our estimates of the NRP for sugar ignore two factors, the first of which indicates that our estimates tend to understate the true NRP, while the second goes in the opposite direction. The first factor is that our estimates make no adjustment for the relatively low polarity (high level of impurities) of planta- tion white. The offsetting factor is the neglect of the cost of bleaching to obtain plantation white. Sugar industry experts have suggested that the low polarity Indonesia 177 Figure 4.2. Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Sugar, Indonesia, 1971­2005 Source: Fane and Warr 2007. effect is probably more important than the effects of bleaching, but that the dif- ference is small. Soybeans Until 1996, the government protected soybean growers by assigning Bulog a monopoly on imports. Since 1996, soybean imports have been unrestricted, and the tariff is currently zero (figure 4.3). The excess of the domestic price of soy- beans over the border price was reduced in 1988, when a local soybean crushing plant run by PT Sarpindo Industri began to operate. However, Bulog prevented the domestic price of beans from falling as rapidly as the world price in 1988­94; and Sarpindo was protected by a local-content scheme that required domestic feed mills to source at least 20 percent of their total usage of soybean meal from local supplies, which meant Sarpindo, the only local supplier of soybean meal. The high cost of feed inhibited the growth of the increasingly powerful poultry indus- try. In 1996, the local-content scheme was abandoned, and Sarpindo was allowed to go out of business. 178 Distortions to Agricultural Incentives in Asia Figure 4.3. Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Soybeans, Indonesia, 1970­2005 Source: Fane and Warr 2007. Maize Maize was a substantial net export industry in the 1970s, but maize subsequently became a net import item. This transition coincided with a movement from neg- ative protection during the export phase to a small amount of positive protection since the early 1980s (figure 4.4). The tariff on imports of processed maize in the form of pellets and flour is currently only 5 percent, but, during the presidency of Megawati Sukarnoputri (2001­04), Rini Suwandi, the minister of trade (sup- ported by Bulog), created import licenses that restricted imports, raising average NRPs well above 5 percent. The rents created by this measure accrue to members of the maize importers association. Other import-competing agriculture: tariffs and tariff surcharges The growth in protection during the 1970s and early 1980s and the reduction in protection in the late 1980s and 1990s were mainly achieved by changing the rates of import tariff surcharges (bea masuk terbahan) rather than the rates of the Indonesia 179 Figure 4.4. Border and Domestic Prices of Import-Competing Products Relative to the GDP Deflator, Maize, Indonesia, 1969­2005 Source: Fane and Warr 2007. import tariff (bea masuk).4 In terms of their economic effects, the surcharges were equivalent to tariffs, but, unlike tariffs, the rates of the surcharges could be changed by administrative decree without the need to amend the law. The rates of import duty shown in appendix table A4 in Fane and Warr (2007) are the com- bined rates of tariffs, plus tariff surcharges. A comparison of the changes in the tariffs and the changes in the tariff surcharges in appendix tables A5 and A6 in Fane and Warr (2007) shows that much of the growth of protection in 1974­79 and almost all of the much larger increases in protection in 1979­85 were achieved by raising tariff surcharges rather than tariffs. When protection was reduced in 1985­89, about half the reduction was obtained by largely eliminating tariff surcharges, which were negligible by 1989, but had been an important part of total import duties in 1985. By 1994, tariff surcharges had been totally abol- ished, and, since then, there has been no need to distinguish between tariffs and total import duties. The import duties on food processing have always been higher than those on agriculture. In every year, the average import tax rate on food processing alone 180 Distortions to Agricultural Incentives in Asia (Harmonized Commodity Description and Coding System, 15­24) is higher than the corresponding average rate on the entire agriculture and food processing products category (1­24). Vegetables and flowers, particularly orchids, have always been the most highly protected of the industries within agricultural cate- gories 1­14. Among the more traditional agricultural subsectors, livestock and estate crops have always received relatively significant protection from imports. However, whereas livestock is mainly import-competing (but also partly nontraded), many estate crops are mainly export-competing. In these cases, of which coffee, tea, and spices are major examples, there is a great deal of water in the tariffs. In the 1970s, the total import tax rates on tea, coffee, vanilla, cinnamon, nutmeg, and ginger were 70 percent; by 1985, this rate had increased to 100 percent. However, by 1989, the total import duty rates on all these products had been reduced to 30 percent, and, by the mid-1990s, they had fallen to 5 percent. The import duty rates on food crops have generally been relatively low, at least in the period before 2000. However, these rates understate the extent of the pro- tection for food crops for two reasons: food crop producers have received input subsidies, and food crops have also been protected by NTBs; hence, their separate treatment above. Estate crops: rubber, copra, coffee, and tea Rubber, copra, coffee, and tea are all produced by perennial plants and, in Indonesia, tend to be produced on large estates, although copra is also produced by small- holders. All have been export crops and all have been taxed, but at varying rates. The export volumes of all these commodities have declined since the 1980s. The high rates of export taxation are a significant part of the explanation. Figure 4.5 shows the calculations of the NRPs for each of these four commodities. For rub- ber, the export tax rate has been low, but the data show high rates of export taxa- tion for copra and tea. For coffee, the rate has declined from the high rates prior to the 1990s. The Political Economy of Protection: Do Good Times Produce Bad Policy, and Bad Times, Good Policy? Key characteristics of Indonesia's political environment provide background on the attempts to explain the changes in trade policy summarized above. First, the confidence of the country's elite was boosted because of the economic successes achieved elsewhere in East and Southeast Asia starting in the 1970s. There seemed Indonesia 181 Figure 4.5. Border and Domestic Prices of Export Crops Relative to the GDP Deflator, Indonesia, 1967­2005 a. Rubber b. Copra (Figure continues on the following page.) 182 Distortions to Agricultural Incentives in Asia Figure 4.5. Border and Domestic Prices of Export Crops Relative to the GDP Deflator, Indonesia, 1967­2005 (continued) c. Coffee d. Tea Source: Fane and Warr 2007. Indonesia 183 no fundamental reason why Indonesia should not succeed as well. This confi- dence meant that bolder strategies could be contemplated. This may be contrasted with the timidity and the skepticism with regard to the international trading sys- tem that characterized most of South Asia at the time. Second, Soeharto's political authority within Indonesia was unchallenged until the end of his regime. Even policies that were unpopular, at least initially, could be contemplated if Soeharto considered them necessary. Observers of economic reform in Indonesia have coined the phrase "good times produce bad policy, and bad times produce good policy," where good times means favorable external conditions, and good policies means deregulation and lower barriers to international trade and investment. The phrase describes much of the history of economic reform in Indonesia. The oil price booms of the 1970s were followed by a series of trade-restricting import-substitution policies aimed at protecting at least some of the traded goods industries that could potentially be harmed by the "Dutch disease" effect of the petroleum booms, that is, the decline in the domestic competitiveness of the traded goods industries caused by a rise in the prices of nontraded goods and services relative to traded goods and services (see Corden 1984; Warr 1986). Trade liberalization followed the adverse terms of trade effect of the decline in petroleum prices beginning in the early 1980s. But, while good times-bad policy, bad times-good policy describes the Indonesian experience, it does not provide an explanation. Why have good times produced bad policy, and bad times, good policy in Indonesia? Observers of policy formation under Soeharto have reported on a contest for Soeharto's attention between the technocrats and the nationalists (Hill 2000). At various times, either of these groups might have had ascendancy, which meant that Soeharto was heeding their messages. The technocrats, many of whom were professional economists trained in the United States, favored a market-oriented economy, a strong emphasis on macroeconomic stability, and a relatively open trade policy. This group dominated the Ministry of Finance and Bappenas, the National Development Planning Agency, and had considerable influence on the Bank of Indonesia. The World Bank used its influence directly in support of the technocrats, and the Bank's resources and technical expertise also assisted the technocrats in making their case more convincing. The economic nationalists were more diverse. They included, in particular, the engineers, led by the minister for research and technology, B. J. Habibie, a German-trained engineer with a strong preference for crash-through economic programs based on advanced technology. This group promoted large-scale, capital- intensive projects in aeronautics, shipbuilding, steel, fertilizers, petrochemicals, and other industries. To ensure the profitability of these projects, high rates of protec- tion were advocated on infant industry grounds. This group was influential within 184 Distortions to Agricultural Incentives in Asia Dr. Habibie's own department, as well as at the state-owned petroleum company, Pertamina. A second group of nationalists was represented by the advocates of self-sufficiency in food, particularly rice. This group dominated the Ministry of Agriculture and Bulog (for example, see Warr 1992; Timmer 1996). More general support for policies based on import substitution was concentrated at the Min- istry of Industry. During bad economic times, the technocrats tended to gain Soeharto's atten- tion. During good times, he listened to the nationalists. The central dynamic derived from the role of the external shocks to the Indonesian economy that operated through petroleum prices. During the Soeharto period, petroleum was both a principal source of foreign exchange, through direct oil exports, and a major source of government revenue, through the royalties received by the gov- ernment on the exports. Reduced oil prices implied both balance of payments and budgetary stresses. In addition, the majority of the country's foreign debt was public debt. When the price of oil fell, the fiscal burden of debt servicing became more painful. This increased the influence of the World Bank, whose willingness to extend concessional loans to the country was important directly and also as a signal to other potential foreign lenders. At such times, the govern- ment needed these loans to balance its budget. The only alternative was infla- tionary financing, the consequences of which had been experienced under Soeharto. The increased influence of the World Bank meant greater influence for the technocrats and the policies they advocated. In addition, reduced oil prices meant a reduction in the influence of Pertamina because they lowered Pertam- ina's contribution to government revenue. It also meant greater influence for the Ministry of Finance, which implemented the tax reforms--designed by tech- nocrats and like-minded foreign advisors--that helped make up for the lost oil revenues. In other countries, a deterioration in the terms of trade might be met by exchange controls, import licensing, and other import-substitution policies. In Indonesia, a protectionist response also occurred briefly after the oil price declines of the early 1980s. But it did not last long because it did not address the simultaneous fiscal problem. An example of the tax reforms that emerged from this dynamic was the introduction of a value added tax in 1986 and, at the same time, a reduction in import duties. An import duty (tariff) is equivalent to both a tax on consumption and a subsidy on production set at the same ad valorem rate. The tariff raises positive net revenue because the volume of consumption of an import commodity exceeds the volume of production of that commodity. A value added tax is a tax on consumption alone. Moreover, it is able to raise the same amount of revenue as a tariff, but at a lower rate of tax. This is so because it does Indonesia 185 not expend revenue to subsidize production. Similarly, the switch from NTBs to tariffs generates revenue. NTBs may be thought of as privately levied tariffs that make no contribution to government revenue. A final example is the phasing out of the Bimas scheme that was designed to help rice growers achieve self- sufficiency. The budgetary cost of the fertilizer and pesticide subsidies and subsi- dized lending involved in Bimas became serious because of the fiscal deteriora- tion of the 1980s. During times of reduced petroleum prices, such as the early to mid-1980s, illiberal trade policies were unaffordable in fiscal terms, and this reinforced the argument that trade liberalization would promote improved foreign exchange earnings from non-oil exports. The technocrats then held sway. In contrast, during the euphoria of the 1970s that was induced by high petroleum revenues, the import-substitution schemes advocated by the nationalists seemed affordable and were politically attractive. Then, the nationalists captured Soeharto's attention. The Asian financial crisis of 1997­98 was the worst of times, and it produced the best of economic policies, given that best is being used here to mean that the policies more closely conformed to the laissez-faire advice of neoclassical econo- mists. The Asian crisis also provides the clearest illustration of the causal link between bad times and laissez-faire policies: the reforms that the government introduced in the wake of the crisis were explicitly adopted to meet IMF loan conditions when all other sources of external lending had dried up. Since Soeharto's political demise in 1998 and the subsequent shift to a more democratic form of government, the president no longer holds absolute authority, and policies are therefore no longer determined simply by a contest between the technocrats and nationalists to influence the president. The parliament, a token institution under Soeharto, now has teeth, and the president cannot ignore its will. Populist economic nationalism has tended to dominate in parliament; and this has reduced, but not eliminated, the influence of the technocrats. In addition, the conspicuous reluctance of the major industrialized countries to lower the protection for their own agricultural sectors has weakened the influence of the technocrats who argue against Indonesian restrictions on trade. The greater protection for the rice and sugar subsectors that followed the end of the IMF pro- gram was a direct consequence of these political changes. Rice self-sufficiency and protection for farmers are both politically attractive goals in Indonesia, and, in the public imagination, both are strongly associated with the national interest. Protec- tion for the rice industry is supported by all major political parties. Because of democracy, rice and sugar farmers therefore now receive more protection from imports than they did under Soeharto. 186 Distortions to Agricultural Incentives in Asia Imputed Protection at the Farm Level The discussion above about protection rates has focused on the effects of policy interventions at the wholesale market level. In this section, we extend the analysis to the way protection (or its opposite) at the wholesale level produces price effects at the farm level. Theory One of the intentions of agricultural protection policy is to influence prices at the farm level. But the goods produced directly by farmers seldom enter international trade. The raw commodities produced by farmers are generally nontraded, whereas the commodities that enter international trade are the processed or par- tially processed versions of these raw products. Between the nontraded raw prod- uct produced by the farmer and the traded processed commodity that enters international trade, there may be several steps: transport, storage, milling, pro- cessing, repackaging, and so on. The significance of this is that border protection policy operates directly on the goods that actually enter international trade either as exports or imports, not on the raw commodities produced by farmers. Protection at the farm level is there- fore a derived effect. It depends on the extent to which policies applied to trade in processed agricultural goods induce changes in prices that are then transmitted to the prices actually faced by farmers. The question thus arises: to what extent do price changes at the wholesale level that are induced by protection policy affect the prices actually received by farmers for the raw products they sell. We construct a simple econometric model to investigate this issue. We use the notational convention that uppercase roman letters (such as X) denote levels in the values of variables and lowercase roman letters (such as x) denote the natural logarithms of the variables, thus: x ln X. Protection at the wholesale level is therefore defined as: PWit P*it(1 TW), it (4.1) where PW denotes the level of the wholesale price of commodity i at time t; Pit* is it the corresponding border price expressed in the domestic currency and adjusted for the handling costs that, in the case of an import, must be taken into account in shifting the cost, insurance, and freight price of the commodity to the domestic wholesale price or, in the case of an export, in shifting the wholesale price of the commodity to the free on board price. The NRP at wholesale is given as TW. In this it discussion, both the border price and the NRP are treated as exogenous variables. The border price is determined by world markets, and the country involved is presumed to be a price taker, that is, the country is presumed to be unable to Indonesia 187 influence the international price. The NRP is determined by the government's protection policy. The farmgate price of a raw material is denoted by PFit, and its logarithm, pFit, is related to the logarithm of the wholesale price by: pFit ai bipWit uit, (4.2) where ai and bi are coefficients, and uit is a random error term. The coefficient bi is the pass-through or transmission elasticity. The estimated values of the coeffi- cients ai and bi are denoted by a^i and b^i, respectively. The econometric estimation of these parameters is discussed below. The estimated coefficients are used as follows. We estimate the logarithm of the farm price that would obtain in the absence of any protection as: p^it F* a^i b^ipW*, it (4.3) where pW* is the estimated value of the wholesale price that would obtain in the it absence of protection, pWit * ln PWit . This is then compared with the estimated * value of the wholesale price in the presence of protection: p^it F a^i b^ipWit . (4.4) We denote the antilogs of p^it and p^it by P^it and P^it , respectively, and then esti- F F* F F mate the NRP at the farm level as: T^it F (P^it F P^it ) P^it. F* F (4.5) It is important to observe that the value of the protection-inclusive farm price used in these calculations is the level estimated in the econometric model (equa- tion 4.4) rather than the actual price given by the raw data. This is because our intention is to use the model to estimate the change in the farmgate price caused by protection at the wholesale level. Thus, both the protection-inclusive price and the protection-exclusive price used in equation (4.5) are the predicted values obtained from the model. The implied NRP at the farm level may be related to the NRP at the wholesale level as follows. Substituting P^it F A^ i(PW)b^i and P^ F* it it A^ i(PW*)b^i into equation it (4.5), where A^ i is the antilog of a^i, then rearranging, and using equation (4.1), we obtain the simple expression: T^it F (1 TWit )b^i 1. (4.6) Obviously, if Tit W 0, then T^it F 0 regardless of the value of b^i. Similarly, if b^i 0, then T^it F 0 regardless of the value of Tit . Also, if b^i W 1, then T^it F Tit . W It may readily be seen that, if Tit W 0, then T^ F Tit as b^i W 1, but if Tit W 0, T^ F TW as b^i it 1. If TW it 0, then T^ F TW as b^i it 1, but T^ F TW as b^i it 1. 188 Distortions to Agricultural Incentives in Asia Econometric application The purpose of the econometric analysis is to estimate the parameter b^i for each commodity.5 For each commodity, we conduct the analysis using time series price data in which each variable is expressed in logarithms, and each is deflated by the GDP deflator for Indonesia: the farmgate price (LFP), the wholesale price (LWP), and the log of the international price, adjusted by the nominal exchange rate and transport and handling costs (LIP). The data extend from 1976 to 2001. The seven commodities for which these data are available are rice, maize, soybeans, sugar, rubber, coffee, and tea. We first tested each of the series (each deflated by the GDP deflator) for the existence of a unit root. For rice, the null hypothesis of a unit root was rejected for all three price series (recalling that they are real, not nominal, price series using the GDP deflator) at the 10 percent level of significance. The price series are thus considered stationary. For other commodities, the results are more mixed. For maize, the null hypothesis of a unit root could not be rejected for farm prices (LFP), but was strongly rejected for the other two price series. For soybeans, the null hypothesis of a unit root could not be rejected for the wholesale price series (LWP), but was rejected at the 10 percent level for the other two series. For sugar, the null hypothesis of a unit root could not be rejected for any of the three series, especially the farm price series (LFP). For rubber, coffee, and tea, the results were similar: the null hypothesis of a unit root marginally failed to be rejected for the farm price series (LFP), but was rejected for the other two series. We first produced ordinary least squares estimates of equation (4.2). In most cases, autocorrelation was a problem, and an AR(1) correction term was included to eliminate it, which was accomplished effectively. The ordinary least squares estimates assume that LFP is endogenous and LWP is exogenous. These assump- tions were tested using Hausman's endogeneity test, although the test has low power when the number of data points is small, as in this case. For each commod- ity, the null hypothesis that LWP was (weakly) exogenous to LFP failed to be rejected, confirming the validity of the ordinary least squares estimates. Reverse Hausman's tests were also conducted, and the null hypothesis that LFP was exoge- nous to LWP was rejected in the cases of maize, sugar, rubber, coffee, and tea. This hypothesis marginally failed to be rejected for rice and soybeans. The results roughly support the validity of the use of the ordinary least squares framework to estimate the transmission elasticity from LWP to LFP, while treating LWP as exogenous. Usable estimates were produced for five commodities: rice, soybeans, sugar, rubber, and tea. The estimated elasticity had the expected positive sign and was significantly different from zero, and the estimated equation performed well. Indonesia 189 Table 4.5. Estimates of Transmission Elasticities from Wholesale to Farm Prices, Indonesia Commodity Estimated elasticity t-statistic Rice 0.73 (5.24) Soybeans 0.53 (3.17) Sugar 0.61 (2.29) Rubber 0.44 (2.60) Tea 0.26 (2.65) Source: Author calculations. Note: The table has been created using the data and methodology discussed in the text. The estimates shown relate to the parameter bi in equation (4.2). Table 4.5 summarizes the estimates. For maize and coffee, the estimated elasticity was not significantly different from zero, and the estimated equation performed poorly. It is often asserted that the commodity price changes that, because of pro- tection or international price movements, arise at the wholesale level are pre- vented by middlemen from being transmitted to farmers. This hypothesis is rejected by the Indonesian data, at least for the five commodities listed above. The transmission elasticities are not zero. Economists often assume that the transmis- sion elasticities are unity. However, the estimated values are generally less than unity and lie between 0.2 and 0.8. The lower values are obtained in the case of rubber and tea, which are perennial crops requiring high processing costs. The other values all exceed 0.5. It is likely that the true transmission elasticities change over time, but the limited data available for this exercise made the assumption necessary that the true values are constant. Estimation of protection at the farm level Given the estimated value of the transmission elasticity, equation (4.6) is used, together with the estimated NRPs at wholesale discussed above, to produce esti- mates of imputed NRPs at the farm level. Because usable estimates of the trans- mission elasticity could not be obtained for three commodities (maize, coffee, and copra), the estimated values for rice, tea, and rubber, respectively, were used as proxies for the true elasticities of these commodities. The imputed farm NRPs are somewhat lower in absolute value than the NRPs at wholesale because the trans- mission elasticities lie between zero and unity, but, given the assumption of con- stant transmission elasticities, they track the pattern of the wholesale results closely. 190 Distortions to Agricultural Incentives in Asia Aggregate Measures of Assistance In this section, we calculate aggregate measures of rates of protection using the information assembled from the preceding analysis and following, as much as possible, the methodology outlined in Anderson et al. (2008) and appendix A. The annual calculations reported fluctuate somewhat from year to year because inter- national and domestic price changes alter the protective effects of all instruments of protection except ad valorem tariffs. In addition, the time taken for domestic prices to adjust to international price changes means that annual data on price differences indicate some variation from one year to the next. The calculations of NRPs at the wholesale and farm levels described in the pre- vious section are used to calculate nominal rates of assistance (NRAs) at the farm level that take account of the assistance for fertilizer input use, in addition to out- put price distortions. Thus, the NRA for a particular commodity is calculated as the NRA on the output of the commodity, plus the product of the cost share of fertilizers in the production of the commodity and the nominal rate of subsidy for fertilizer use (which was up to two-thirds of the price of fertilizer in the early 1970s, but which has declined since then and, in recent years, has been only one- fifth of this price). The aggregate NRA therefore exceeds the NRAs on output for each commodity that uses fertilizer as an input. The calculations of the NRAs confirm that, in 2000­04, import-competing commodities were significantly protected, notably rice and especially sugar (table 4.6). The rates of assistance for these two commodities have increased sig- nificantly relative to the rates in the 1990s, whereas the assistance for maize and soybean producers has fallen. Tea is still moderately taxed, but export commodi- ties such as rubber, copra, and coffee are either only lightly taxed or only slightly assisted today, having been taxed--sometimes significantly--prior to the late 1990s. The average NRA for import-competing farm products is always above the average NRA for exportables, and the extent of this antitrade bias within the agri- cultural sector has not diminished over time (figure 4.6 and table 4.7). Nor has the dispersion of the NRAs of covered products declined as measured by the annual standard deviation around the weighted mean (table 4.6), which means that the inefficiency of the use of resources across industries within the sector remains nontrivial. Finally, the relative rate of assistance (RRA) to agriculture is a function of the difference between the NRA in tradable agriculture and the NRA in nonagricul- tural tradables such as manufactures, but also nonfarm primary fishing, forestry, and mining products.6 Our RRA estimates show that, before the Asian crisis of the late 1990s, agriculture was effectively taxed by between one-quarter and one-third, but, shortly after the crisis, it was a slightly net subsidized sector (table 4.7 and fig- ure 4.7). Because we have erred on the side of understating rates of manufacturing Table 4.6. NRAs for Covered Agricultural Products, Including Fertilizer Use Subsidies, Indonesia, 1970­2004 (percent) Indicator, product 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Exportablesa 3.3 0.3 7.0 16.5 24.6 17.2 3.0 Coffee 7.1 3.7 8.6 2.2 0.5 2.3 3.0 Tea 6.3 1.9 1.8 2.3 2.5 13.9 15.5 Coconuts 5.9 2.2 6.1 22.0 45.6 29.4 8.1 Rubber 15.2 3.4 16.2 20.5 31.9 37.0 16.7 Palm oil 14.5 9.2 22.2 1.1 11.9 18.3 3.8 Import-competing productsa 3.6 16.5 19.5 5.1 0.7 5.8 24.7 Rice -- 13.9 7.5 0.9 8.7 13.0 18.7 Maize 15.4 10.2 18.6 21.9 22.5 24.6 10.8 Soybeans 5.9 31.8 49.0 17.0 17.7 17.5 1.2 Sugar 2.1 23.5 53.8 8.5 3.9 11.3 49.4 Poultryb 72.8 144.3 147.5 86.8 94.9 87.9 99.8 All covered productsa 3.9 11.1 12.2 0.3 5.5 9.1 16.4 Dispersion, covered productsc 27.6 49.4 53.6 35.0 40.5 49.0 36.3 Coverage, at undistorted prices 65 68 64 61 64 63 59 Source: Fane and Warr 2007. Note: The output-subsidy equivalent of the subsidy for fertilizer input is incorporated in the NRA for each crop. It is estimated by multiplying the input subsidy by the share of fertilizer in the cost of production of each product and adding it to the output NRA. -- no data are available. a. Weighted averages; the weights are based on the unassisted value of production. b. The first and last periods refer to 1971­74 and 2000­03, respectively. c. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of covered products. 191 192 Distortions to Agricultural Incentives in Asia Figure 4.6. NRAs for Exportable, Import-Competing, and All Agricultural Products, Indonesia, 1970­2004 Source: Fane and Warr 2007. protection prior to 1987, better estimates of manufacturing protection during this period would show larger negative RRAs prior to the 1990s and would thereby reinforce rather than undermine the above finding. Conclusions and the Prospects for Reform Having previously taxed agriculture relative to the nonfarm sectors during the postindependence period, Indonesia's trade policies have, on average, not taxed agriculture since around 2000. This change occurred following the Asian crisis of the late 1990s. The switch took the form of (a) increases in the protection of the import-competing commodities sugar and rice; (b) declines in the taxation of agricultural exports, especially rubber and copra; and (c) declines in manufactur- ing protection. The transition to a more democratic form of government has weakened the influence of the country's technocrats, who have generally favored liberalized trade policies. Greater protection among some key agricultural com- modities has been a consequence. Assistance for agriculture primarily takes the form of protection for the import-competing sugar and rice subsectors. Other output subsectors receive virtually no direct assistance. Subsidies for fertilizers and other inputs were once an indirect source of assistance for agriculture, but the nominal rates of these subsi- dies have declined. Table 4.7. NRAs in Agriculture Relative to Nonagricultural Industries, Indonesia, 1970­2004 (percent) Indicator 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Covered productsa 3.9 11.1 12.2 0.3 5.5 9.1 16.4 Noncovered products 2.3 5.4 4.2 3.8 8.4 7.7 7.1 All agricultural productsa 3.4 9.3 9.2 1.7 6.6 8.6 12.0 Non-product-specific assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total agricultural NRAb 3.4 9.3 9.2 1.7 6.6 8.6 12.0 Trade bias indexc 0.03 0.14 0.21 0.20 0.24 0.12 0.22 NRA, all agricultural tradables 3.8 10.4 10.5 1.9 7.5 9.7 13.9 NRA, all nonagricultural tradables 27.7 27.7 27.7 26.5 17.6 10.6 8.1 RRAd 24.7 13.6 13.5 22.5 21.3 18.3 5.4 Source: Fane and Warr 2007. a. Including product-specific input subsidies. b. Including product-specific input subsidies and non-product-specific assistance. Represents total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. c. The trade bias index is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. d. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. 193 194 Distortions to Agricultural Incentives in Asia Figure 4.7. NRAs for Agricultural and Nonagricultural Tradables and the RRA, Indonesia, 1970­2004 Source: Fane and Warr 2007. Note: For the definition of the RRA, see table 4.7, note d. The political explanations for the protection of sugar and the protection of rice are quite different. The protection of sugar is a consequence of the political power of the highly concentrated refining industry, including the state-owned compo- nent of this industry, which is closely linked with large-scale sugar plantations. In contrast, the farm production of rice (paddy) is dominated by small-scale farm producers. The rice milling industry is much more concentrated and much more well organized, however, and this is relevant because rice imports compete with milled rice rather than the raw, unmilled product (paddy) produced by the farmers. The political power of rice millers has been an important source of sup- port for the protection of the rice industry. The enhanced political power of the Indonesian parliament since the upheavals induced by the Asian crisis, together with the economic nationalism that dominates among members of parliament, has strengthened the support for the protection of the rice industry. Since 2004, rice imports have officially been banned. In part, this policy has reflected the mistaken claim, advanced by supporters of rice industry protection, that restricting rice imports reduces poverty. A general equilibrium analysis presented in Warr (2005) shows that the policy is more likely to have increased poverty within both rural and urban areas because the poverty-increasing effects of raising the consumer price of rice far exceed the poverty-reducing effects of raising the producer price. Indonesia 195 Given the politics at work, trade liberalization in the country's sugar and rice industries seems unlikely in the foreseeable future. Indeed, rising protectionism seems a more likely outcome, and this might conceivably extend to industries other than sugar and rice. Notes 1. Soeharto's New Order began in March 1966, although Soekarno, Soeharto's predecessor, remained nominal president for 12 more months. The statement about rice self-sufficiency was reportedly made during Soeharto's visit to the headquarters, in Rome, of the Food and Agriculture Organization of the United Nations, in 1985, when Indonesia's rice imports stood (briefly) at zero. 2. The border price of rice in chart a, figure 4.1 has been converted to make it as nearly comparable as possible to the wholesale price. The free on board price was adjusted to the cost, insurance, and freight level by adding freight and insurance costs; the resulting price was then adjusted to the whole- sale level by adding margins to allow for the estimated handling, warehousing, and interest costs. 3. Gross indicates that each hectare that is harvested twice in a single year is counted as 2 hectares. 4. The term tariff surcharge is a misnomer in the sense that the base to which the rates of the tariff surcharge applied was not the tariff, but the border value (cost, insurance, and freight) of the imports subject to the tariff surcharge. For example, in 1985, in the case of live animals (other than pure bred), the tariff was 30 percent, and the tariff surcharge was 15 percent, giving a total import duty of 45 per- cent of the border value (cost, insurance, and freight). 5. The results are summarized here. Details of the econometric analysis are available upon request. 6. The NRAs for nonagricultural tradables are estimated mainly on the basis of the effective rates of protection for manufacturing in Fane and Condon (1996), who estimated these rates for 1987 and 1995 at 48 and 20 percent, respectively. These authors also projected the corresponding effective rate for 2003 at 13 percent based on the May 1995 tariff reduction package, which was to be implemented by 2003 and which was, in fact, largely implemented. For all years before 1987, we have used the 1987 values even though some tariff reduction occurred during the few years before 1987. For 1987­95 and 1995­2003, we have interpolated linearly. For 2004, we have used the 2003 value. (As noted above, the focus of this discussion is the identification of broad long-run trends in the structure of protection rather than annual fluctuations.) References Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Booth, A. 1988. Agricultural Development in Indonesia. Sydney: Allen and Unwin. BPS (Badan Pusat Statistik, Statistics Indonesia). 1971. Input-Output Tables, Indonesia. Jakarta: BPS. ------. 1980. Input-Output Tables of Indonesia. Jakarta: BPS. ------. 1990. Input-Output Tables of Indonesia. Jakarta: BPS. ------. 2000. Input-Output Tables of Indonesia. Jakarta: BPS. Corden, W. M. 1984. "Booming Sector and Dutch Disease Economics: A Survey." Oxford Economic Papers 36 (3): 359­80. Fane, G., and T. Condon. 1996. "Trade Reform in Indonesia, 1987­1995." Bulletin of Indonesian Economic Studies 32 (3): 33­54. Fane, G., and P. Warr. 2007. "Distortions to Agricultural Incentives in Indonesia." Agricultural Distor- tions Working Paper 24, World Bank, Washington, DC. Hill, H. 2000. The Indonesian Economy. 2nd ed. Cambridge: Cambridge University Press. 196 Distortions to Agricultural Incentives in Asia Timmer, C. P. 1996. "Does Bulog Stabilize Rice Prices in Indonesia? Should it Try?" Bulletin of Indonesian Economic Studies 32 (2): 45­74. Warr, P. G. 1986. "Indonesia's Other Dutch Disease: Economic Effects of the Petroleum Boom." In Natural Resources and the Macroeconomy, ed. J. P. Neary and S. van Wijnbergen, 288­320. Oxford: Basil Blackwell. ------. 1992. "Comparative Advantage and Protection in Indonesia 1." Bulletin of Indonesian Economic Studies 28 (3): 41­70. ------. 2005. "Food Policy and Poverty in Indonesia: A General Equilibrium Analysis." Australian Journal of Agricultural and Resource Economics 49 (4): 429­51. World Development Indicators Database. World Bank. http://go.worldbank.org/B53SONGPA0 (accessed May 2008). 5 MALAYSIA Prema-Chandra Athukorala and Wai-Heng Loke Malaysia is notable among developing countries for its long-standing commit- ment to the maintenance of a relatively open trade policy regime. It has never relied heavily on quantitative restrictions or other forms of nontariff protection, and its tariffs on domestic manufacturing and agriculture have been low relative to similar tariffs in other developing countries. In the first half of the 1980s--as part of a government-led strategy of heavy industrialization--and in the immedi- ate aftermath of the 1997­98 economic crisis, tariffs on some manufactured goods were substantially increased, and some goods were brought under quantitative restrictions to support selected domestic industries in the face of a massive domestic economic contraction. However, these measures were eventually reversed, and additional tariff cuts were undertaken in the ensuing years. Unlike its counterparts in other countries, the Malaysian government has con- sistently avoided exerting direct influence on export prices or the direct procure- ment of agricultural output through government marketing boards. Export duties on the two major primary commodities, rubber and palm oil, were a major source of government revenue until the mid-1980s. However, export taxes were periodi- cally adjusted in line with world price trends to keep producer prices relatively stable. The government also plowed a significant share of this revenue into well- designed and efficiently implemented replanting schemes and productivity- enhancing research in these industries. Over the past two decades, export duties on rubber and palm oil have decreased drastically. However, these traditional export industries have been under persistent domestic cost pressure, which has arisen from the rapid structural transformation of the economy through success- ful global integration. 197 198 Distortions to Agricultural Incentives in Asia The production of paddy rice--the staple food of the country and the princi- pal domestic food crop--stands out as the single most heavily assisted economic activity. During the immediate postwar years, assistance for this crop was intro- duced by the colonial government on the grounds of food security. After inde- pendence, particularly after the launch, in 1970, of the New Economic Policy (Dasar Ekonomi Baru), a sweeping affirmative action policy package aimed at increasing the share of certain ethnic groups in the economy, rice became an increasingly sensitive political crop. While the achievement of food self-suffi- ciency has continued to represent a moving target in successive development plans, protection for rice farmers, who are concentrated in relatively economically backward states, remains the overwhelmingly predominant reason behind protec- tion. Producers of subsidiary crops, horticultural products, livestock, and fishing products have been operating under a virtually free trade regime throughout, with the exception of quantitative restrictions on the importation of round cabbages. High-value (processed) food products have emerged as an important dynamic export product over the past two decades. This chapter has two main purposes: first, to provide an analytical narrative of the nature and evolution of trade in Malaysia and the related and accompanying policies that have had an impact on domestic agriculture (with a focus on the underlying political economy) and, second, to examine the degree of and chang- ing patterns in the policy distortions affecting the incentives in domestic agricul- ture, including both direct (sector-specific) incentives and indirect incentives emanating from economy-wide policies. We also aim to inform the debate in Malaysia on the future direction of national development policy as it relates to domestic agriculture. The analysis is undertaken against the backdrop of the ongoing process of rapid structural transformation of the country's economy over the past three decades. As an integral part of the analysis, we attempt to delineate the implications--over and above the implications of policy-induced incentives-- of the process of structural transformation for the long-term viability of tradi- tional plantation crops and the new opportunities for agricultural output expan- sion in the subsidiary food-crop subsector. Our study covers the period from 1960 (the earliest year for which reliable data are available after independence in 1957) to 2004. We emphasize four major prod- ucts: rubber, palm oil, cocoa, and rice. The first three are exportable products, while the fourth, the main food staple, is an import-competing product. Together, the four products accounted for three-fifths to four-fifths of the value of total agricultural gross domestic product (GDP) over the period. Because of the paucity of data, we are unable to cover other commodities, but virtually all other agricultural products have been facing free trade conditions throughout the period we study. Malaysia 199 The structure of the chapter is as follows. The next section provides an overview of growth and structural change in the economy during the postinde- pendence era (since 1957), with an emphasis on the relative importance of the agricultural sector and the trends and compositional shifts in agricultural output and trade. This is followed by a survey of the evolution of agricultural trade policy during the era since 1957 against the backdrop of overall national development strategy and macroeconomic policy. We pay particular attention to the political economy considerations that underpin policy directions. In the subsequent sec- tion, we provide estimates of the extent of and patterns in the direct and indirect distortions in the incentives faced by domestic agriculture. The estimates are based on indicators that have been specifically constructed for our study. The final section contains a summary of the key findings and policy inferences. Agriculture in the Malaysian Economy To aid an understanding of the policy environment in Malaysia, we explain the growth trends in the economy and the patterns of structural change intersec- torally and within the agricultural sector. Economic growth trends At independence in 1957, Malaysia was a classic agrarian economy in which agri- culture and the mineral industry were concentrated in primary production for export (Levin 1960). Natural rubber directly accounted for 25 percent of GDP, while the second largest export, tin, accounted for 5 percent (Meerman 1979). In addition, a host of service activities that embraced trade, transport, and finance was dependent on the export sector. Production patterns exhibited only limited changes until about the mid-1970s. The structural changes since then have been dramatic, however. In particular, beginning in the late 1980s, there has been an expansion in export-oriented manufacturing and related activities in the modern sector. The share of agriculture in GDP had declined to 19 percent by the late 1980s and then plummeted to 8.5 percent in 2005 (table 5.1 and figure 5.1). In the early 1970s, agriculture directly absorbed about half the total labor employed in the country. This share had declined to 13.7 percent by 2005. The contraction in the relative importance of agriculture in total labor absorption was particularly marked over the two decades before 2005 (table 5.1). Despite this dra- matic structural transition, the promotion of employment opportunities in the rural economy, which is roughly equivalent to the agricultural sector, has contin- ued to be an important focus of the government's successive 10-year development plans. This is so particularly because of the delicate ethnic dimension involved: over 200 Distortions to Agricultural Incentives in Asia Table 5.1. Agriculture in the Economy, Malaysia, 1970­2005 (percent) Real annual average growth Share of agriculture Year In GDP In agriculture In GDP In employment 1970­74 2.3 3.4 25.5 50.9 1975­79 7.3 5.2 23.3 46.4 1980­84 6.6 3.4 20.4 39.5 1985­89 4.8 4.3 19.1 32.4 1990­94 9.3 0.2 15.3 26.9 1995­99 5.2 0.1 10.1 17.9 2000­04 5.2 3.8 8.7 15.0 2005 5.3 2.1 8.5 13.7 Source: Ministry of Finance, various. Figure 5.1. Commodity Composition of Agricultural GDP, Malaysia, 1965­2005 100 90 , 80 GDP % 70 60 averages, 50 agricultural 40 of 30 five-year share 20 10 0 1965­69 1970­74 1975­80 1980­84 1985­89 1990­94 1995­99 2000­05 year rubber palm oil cocoa paddy livestock other Sources: Author compilation based on data of the Economic Planning Unit; Athukorala and Loke 2007. Note: The figure excludes forestry and forestry products. 90 percent of the agricultural labor force belongs to the Malay community. The incidence of poverty in the rural economy is still high. The poverty rate was 11 percent in the rural economy in 2003, while it was only 2.2 percent in the urban sector and 6 percent, on average, at the national level (Economic Planning Unit 2006). Malaysia 201 Over the past three decades or so, the agricultural sector has been under con- stant pressure because of the resource-pull effects of the rapid structural changes in the economy (Barlow and Jayasuriya 1987; Athukorala and Manning 1999). Widening urban-rural wage differentials and an aversion among younger people to engagement in agricultural pursuits has increased rural-to-urban labor migra- tion. This has caused widespread labor shortages in the rural economy and added to the stress on agricultural wages. The area under traditional plantation crops has shrunk in semiurban areas because of the dispersion of industrial centers across Peninsular Malaysia and the resulting higher demand for land for residential and industrial expansion. In the face of severe shortages of local labor, agricultural producers--at first, plantation enterprises and, more recently, smallholder pro- ducers of cash crops and paddy--began to rely increasingly on foreign workers. The estimated foreign labor force in Malaysia increased from around 500,000 in the mid-1980s to 1.8 million (23 percent of the total labor force) by 2003; about half of these workers were in the agricultural sector (Athukorala 2006a). The dependence on foreign labor is particularly high in the rubber industry because of the relatively high labor intensity of cultivation and harvesting. However, the rela- tively more capital-intensive palm oil industry also began to depend more on for- eign workers because fresh fruit bunch-harvesting is conducted manually. Accord- ing to a recent Malayan Agricultural Producers Association survey, 37 percent of the total labor force on private sector plantations in Peninsular Malaysia are for- eign workers, while the share is as high as 80 percent in East Malaysia (Khoo and Chandramohan 2002). Structural changes: plantation (cash) crops Beginning in the 1890s, rubber was the preferred crop of the foreign-owned estate sector during the colonial period (Barlow 1978). Oil palms were first commer- cially planted in 1917 (although palm oil remained a relatively minor crop until the mid-1950s). During the ensuing three decades, palm oil proved to be more profitable than rubber and expanded in the plantation subsector at the expense of rubber. The government policy emphasis in the plantation subsector shifted dra- matically from rubber to palm oil, and the support for the expansion of palm oil production increased when rubber replanting grants--grants to plant oil palms on land still under rubber--began to be distributed. The government played a vigorous role in the expansion of palm oil production by embarking on a large resettlement effort (especially though the Federal Land Development Authority) under which palm oil was the crop of choice (Pletcher 1991).1 The country's suc- cess in promoting palm oil exports was aided by the inappropriate agricultural and economy-wide policies of the traditional palm oil exporting countries in 202 Distortions to Agricultural Incentives in Asia Africa (MacBean 1989; Athukorala 1991). Cocoa gained importance as an alterna- tive crop in the plantation subsector starting around the beginning of the 1970s. However, about the mid-1980s, producers began to contract the area under cocoa cultivation sharply because of the persistently low world prices for cocoa beans. Around the late 1980s, the plantation subsector came under severe strain because of the resource-pull effects of the rapid structural changes in the econ- omy (see elsewhere above). These effects were felt more acutely in the rubber and cocoa industries, where the cultivation and harvesting processes were more labor intensive relative to the palm oil industry. Moreover, the world prices for the two former products were unfavorable relative to the world price of palm oil. At the end of the crop cycle, many plantation enterprises and smallholders replanted some of the land given over to rubber with less labor-intensive crops, particularly oil palms, or used the land for residential and industrial expansion. Many rubber estate companies were facing hard times. They began employing foreign workers to overcome labor shortages, and then they used their technology and their management expertise to invest in neighboring countries (particularly Indonesia and the Philippines) in which wages and land cost less. Smallholders also continued to tap rubber using hired migrant labor, although the higher price elasticity of supply among the smallholders resulted in declining output when output prices fell in the early 1990s (Barlow 1997). In recent years, despite the nat- ural cushion provided by the relative capital intensity of production and the rela- tively favorable price trend, the palm oil subsector has also felt the pinch of the resource-pull effects. Many large plantation enterprises have shifted their invest- ments to neighboring land- and labor-abundant countries. The three main plantation crops are rubber, palm oil, and cocoa. The area under rubber increased from 1.7 million hectares in the early 1960s to 2.0 million hectares in the late 1970s and early 1980s. However, it has declined steadily since then, reaching 1.3 million hectares in 2000­04. The area under oil palm cultiva- tion rose from less than 100,000 hectares in 1960­64 to 3.8 million hectares in 2000­04. The expansion in output was more rapid in both rubber and palm oil than the expansion in area under cultivation. This reflects the widespread use of new high-yielding varieties and improved cultivation practices. Since the late 1990s, the rubber and palm oil industries have benefited from increased world demand, especially growing demand in China, and the resulting favorable price trends. The palm oil industry has benefited additionally in recent years from the tight world supply of edible oils and fats and the expanding demand for biodiesel fuel, which has pushed up palm oil prices. In the case of rubber, the supply response to favorable prices took the form of greater cropping intensity in the context of the persistent decline in cultivated area. In contrast, the area under oil palm cultivation showed some mild positive Malaysia 203 response to favorable prices. In cocoa, both the output and the area under cultiva- tion have declined steadily since the early 1990s. Meanwhile, the yield per hectare in palm oil rose throughout. In rubber, the yield increased notably in the 1960s and 1970s, but this was followed by virtual stagnation. The stability of rubber yields in the face of the decline in the total out- put of rubber occurred largely because of a contraction in the area under cultiva- tion, which suggests that marginal plantations showing poor yields were being abandoned (Athukorala and Loke 2007). The smallholder share of the land under rubber cultivation has always been high. In 1960­2004, it grew from 50 to 93 percent. The relatively capital-intensive palm oil industry was initially dominated by large plantations, although the smallholder share rose from 16 percent in 1970 to around 40 percent after 1990 (Athukorala and Loke 2007).2 The dominance of estate cultivation is an unusual feature of the palm oil industry in Malaysia (and in Indonesia) because the small- holder production of palm oil is much more important in other parts of the world (Pletcher 1991). Structural changes: food crops Rice farming, nearly all wet paddy farming, is the major source of income for rural households in the states of the north and east in Peninsular Malaysia and parts of East Malaysia. At independence, about three-quarters of the native peasant pro- ducers (predominantly rice growers and fishermen) were Malays; about 90 per- cent of the rice growers were Malays; and about one-third of the economically active Malay male population represented the peasant sector (Meerman 1979). This ethnic dimension of rice farming persisted during the ensuing years, making rice a highly sensitive political crop. Nearly half of all peasant cultivators grow some rice. The area under paddy cul- tivation rose from 490,000 hectares in 1960­64 to 750,000 hectares in 1970­74 and has remained near this level since then. Paddy production rose steadily from the early 1960s to the mid-1990s, however, because of increases in yield per hectare starting in the 1970s. Paddy producers were aided significantly by govern- ment policies such as sponsored irrigation schemes that permitted double crop- ping, the introduction of high-yielding varieties, the consolidation of paddy small- holdings through group farming in the eight granary areas, and direct assistance to farmers through price supports and credit and fertilizer subsidies (see below). By the end of the eighth plan period in 2005, almost all farming operations in major paddy growing areas had been fully mechanized. As a result, the labor input per hectare declined from 47 to 15 worker days in 1995­2000 (Economic Planning Unit 2001). At independence in 1957, about 45 percent of all rice consumed in 204 Distortions to Agricultural Incentives in Asia Malaysia was imported, but, since the mid-1970s, around 90 percent of the rice consumed has been produced in the country. Beginning in the mid-1990s, rice production stagnated. This has been followed by a mild, but steady decline in recent years despite the constant rise in yields per hectare (Athukorala and Loke 2007). As in the case of plantation crops, paddy farming has been under pressure because of labor shortages arising from rural-to- urban labor migration and the aging of the farming community (Ahmad and Tawang 1999). It seems that massive government support has been insufficient to maintain paddy farming in the context of the enormous pressure emanating from the ongoing structural adjustment in the economy at large. A noteworthy development in nonplantation agriculture over the past 15 years has been the rise in the output of subsidiary food products, such as fruits, vegeta- bles, fish, and livestock, which has been growing at a more rapid rate than paddy output (Economic Planning Unit 2006). While there has been some increase in the exports of these high-value food products, the expansion in production has so far gone predominantly to satisfying domestic demand. This is because of the rapid income growth in the modern sector of the economy and the high income elasticity of demand for high-value food products relative to rice. The importance of the subsidiary food subsector is bound to rise rapidly in years to come. Figure 5.1 provides data on the composition of the value added by agricultural production in the economy. The dominance of the plantation crop subsector in agriculture declined steadily, from 73 percent in the early 1970s to 53 percent in 2000­05. Within this subsector, there has been a noticeable shift in the composi- tion of output from rubber toward palm oil. Cocoa accounted for around 10 per- cent of total agricultural GDP from the mid-1970s to the early 1990s, but this share declined thereafter, reaching negligible levels by the end of the 1990s. The combined share of food crops, which remained around 28 percent in the 1960s and 1970s, rose sharply during the ensuing years, reaching 47 percent in 2000­05. Within this subsector, the relative importance of paddy has declined appreciably over the past two decades. This reflects the compositional shift toward livestock, fish, and other subsidiary food crops, mainly fruits and vegetables. Agricultural trade The dramatic shifts in the structure of domestic production were closely mirrored in the patterns of export. The combined share of agricultural products in total merchandise exports declined from 39 percent in 1970­74 to 8 percent in 2000­04 (table 5.2). (The respective shares are 58 and 10 percent if timber and wood exports are included.) Among agricultural products (excluding timber and wood), the share of rubber declined from 62 to slightly less than 10 percent over the same years; this was offset by an increase in the share of palm oil from 24 to 42 percent. Malaysia 205 Table 5.2. Product Shares, Agricultural Exports, Malaysia, 1970­2004 (percent) Indicator 1970­74 1975­79 1985­89 1995­99 2000­04 Agriculture in merchandise exports Total 58 54 36 13 10 Excluding timber and wood 39 38 24 11 8 Composition, agricultural exportsa Rubber 62.4 52.9 33.7 12.9 9.7 Palm oil 23.9 27.5 31.9 44.1 41.6 Cocoa 2.1 2.3 6.4 2.5 2.6 Spices 2.2 2.1 1.5 1.1 0.8 Processed food 8.7 8.5 15.6 26.2 31.6 World market share Rubber 62 57 37 20 19 Palm oil 70 77 75 64 55 Cocoa 3 3 11 4 4 Sources: Ministry of Finance, various; Bank Negara Malaysia, various. a. Excluding timber and wood. The share of Malaysia in the total world exports of natural rubber fell from 62 to around 20 percent between the late 1970s and 2004, when Malaysia was the third most important exporter of rubber in the world after Thailand and Indonesia (Athukorala and Loke 2007). In world crude palm oil exports, Malaysia was the largest exporter until recently, accounting for a share between 65 and 80 percent from the mid-1970s to the late 1990s. Indonesia has since become the world's largest exporter; it now accounts for over 50 percent of total world exports (not shown in table 5.2). This is the result of massive investment in relocation by Malaysian plantation companies in the face of mounting domestic cost pressures. A noteworthy development in the structure of exports over the past two decades has been the emergence of processed food as a dynamic export line. The average annual growth rate of processed food exports rose from 5 to 19 percent between 1985­99 and 2000­04. The share of processed food in total agricultural exports increased from 9 percent in 1970­74 to 16 percent in 1985­89 and 32 per- cent in 2000­04 (table 5.2). The rapid expansion of processed food compared with traditional food products (coffee, tea, sugar, cocoa, and so on) has been a universal phenomenon in world trade over the past two decades (Athukorala and Jayasuriya 2003).3 The recent performance of Malaysia in this lucrative growth area, while impressive relative to past performance, has lagged behind that of 206 Distortions to Agricultural Incentives in Asia many other counties (particularly Thailand) with similar agricultural resource endowments (Athukorala 2006a). An interesting issue that deserves more analysis revolves around whether the country's highly interventionist paddy sector sup- port policy has constrained a shift in agricultural resources to these new, more dynamic product lines. Policy Context The development prospects of Malaysia (then the Federation of Malaya) at inde- pendence in 1957 were mixed at best.4 On the positive side, the country's per capita income was on a par with incomes in Hong Kong, China and Taiwan, China and higher than incomes elsewhere in East Asia, save Japan. Although the rate of population growth was already rapid, the highly favorable ratio of land and other natural resources to total population offered great potential to raise per capita incomes. Moreover, the colonial inheritance included well-developed infrastruc- ture, efficient administrative mechanisms, and a thriving primary export sector with considerable potential for expansion. The mobilization of this development potential to build the economy of the newly independent Malaysia had to be accomplished in the face of challenges posed by a pluralistic society inherited from the colonial past. At the time, ethnic Malays, who accounted for 52 percent of the population, dominated in politics, but were relatively poor and were involved mostly in low-productivity agricultural activities. Ethnic Chinese (37 percent of the population) enjoyed greater eco- nomic power and dominated in the modern sector, but they did not match the ethnic solidarity or political power of the Malays.5 Ethnic divisions weakened the national fabric; the machinery of government was fragile, and the democratic political leadership remained untested. In this context, there was not much room for optimism regarding the development policies that might be expected from the newly elected government (World Bank 1955). The development challenges faced by the country were generally considered more problematic than the challenges being faced in other countries that had also newly emerged from a colonial past, particularly Ghana, India, Kenya, Myanmar, and Pakistan.6 From independence to the mid-1960s, national development policy was gener- ally in line with traditional liberal notions of the limited state. The public command of economic resources in these early years was narrow, and prevailing economic policies were conservative. The policy thrust basically involved contin- uing the colonial open-door policy stance relating to trade and industry, while attempting to redress ethnic and regional economic imbalances through rural development schemes and the provision of social and physical infrastructure (Snodgrass 1980). Malaysia 207 As in many other developing countries, industrialization through import sub- stitution was a key part of the Malaysian development strategy during this period. However, Malaysian policy makers, unlike their counterparts in other countries, did not seek forced industrialization through direct import restrictions and the establishment of state-owned industrial enterprises (Lim 1992).7 Moderate tariff protection was, by and large, the key instrument used to encourage new invest- ment in manufacturing. The average tariff rate in 1965 has been estimated at a mere 13 percent, and few industries enjoyed nominal tariffs above 30 percent. Nontariff barriers were almost nonexistent (Power 1971; Lin 1984). The race riots in Kuala Lumpur in 1969 generated a dramatic shift in develop- ment policy along ethnic lines. The leadership of the ruling National Front con- cluded that severe discrepancies in wealth had to be rapidly eliminated, partly though public sector activity, if the country was to evolve into an integrated com- munity. The basic goals of the leadership were the eradication of poverty through a rise in incomes and the creation of more opportunities for all Malaysians irre- spective of race, as well as a rapid transformation of society to correct economic imbalances and reduce and eventually eliminate the association between race and economic role (Government of Malaysia 1971). However, in the language of the New Economic Policy, a larger share of gross national product among Malays was not to be provided at the expense of the citizens in other ethnic groups. Given the delicate ethnic composition of the ruling coalition, economic equality was to be fostered primarily by raising employment and establishing a mechanism to ensure that a greater share of any newly generated assets would accrue to Malays. The redistribution of existing assets was anathema. Nationalization, land reform, and local or foreign expropriation were not considered in the New Economic Pol- icy (Ness 1967; Snodgrass 1980; Ganguly 2003). Because of the crucial role played by foreign-owned companies in the produc- tion and marketing of plantation crops, the government took care to pursue favorable and unambiguous policies toward direct foreign investment. Transfer- ring a progressively larger share of foreign-owned companies to local ownership was a declared policy objective, but the government consistently declared that the transfer of ownership would occur through formal share trading rather than through arbitrary expropriation (Myint 1984; Pletcher 1991). In the first half of the 1980s, the promotion of heavy industries through direct government involvement was emphasized as part of the Look East policy of Mahathir bin Mohamad, who became prime minister in 1981. The symbol of the selective industrial policy that resulted was the Proton, the Malaysian national car project, a joint venture of DRB-HICOM and the Mitsubishi Corporation, Japan. By 1987, there were 867 public corporate enterprises in the country, more than one-third of which were manufacturing enterprises. Tariffs on a wide range of 208 Distortions to Agricultural Incentives in Asia manufactured goods were substantially increased in the first half of the 1980s as part of the shift toward heavy industrialization. However, there was no significant reliance on quantitative import restrictions (Athukorala and Menon 1999). The economic crisis of 1985­87--caused by a combination of adverse price trends among the country's major export products and the budget deficits arising from the shift toward heavy industrialization--put an end to the government-led push toward heavy industrialization. The policy package assembled to manage the crisis placed greater emphasis on the role of the private sector and fostered more favorable conditions for export-oriented industrialization through the greater participation of foreign direct investment. The structural adjustment reform package that was subsequently introduced sought the gradual privatization and restructuring of publicly owned enterprises. The reforms of the late 1980s also involved significant tariff reductions and the removal of quantitative import restrictions. Some of the tariff increases introduced in the first half of the 1980s were reversed, and additional tariff cuts were implemented as part of the market- oriented reforms. By the early 1990s, public sector ownership in manufacturing was limited to a modest number of politically sensitive ventures in automobile manufacturing (the Proton project) and in the cement, petrochemical, and iron and steel industries. The policy response to the 1997­98 Asian financial crisis involved a departure from the persistent trade liberalization of the previous decade (Athukorala 2002). The 1998 budget relied on higher import duties on automobiles, vans, and motor- cycles. The duties rose from 30­200 to 40­300 percent for completely assembled motor vehicles, from 4­42 to 30­80 percent for unassembled motor vehicles, and from 0­35 to 5­50 percent for construction equipment. In addition, other prod- ucts were brought under nonautomatic import licensing. This included heavy equipment, construction equipment, hot-, cold-, and flat-rolled products of iron or nonalloy steel, ephedrine and ephedrine salts, chemical products, and selected household electrical appliances. The declared purpose of these measures was to bring down the current account deficit; but cushioning local producers, including the national car producer, Proton, against contractions in domestic demand was also a key motivating factor. However, there was no notable retreat from the long- standing commitment to a highly open trade regime. Despite recent tariff increases, the average applied import duty rate--total duty collection as a percentage of total merchandise exports--has declined steadily (Athukorala and Loke 2007). The underlying tariff structure is far from uniform, however. The domestic automobile market is heavily protected through tariff and nontariff measures. At the two-digit level of the Harmonized Commod- ity Description and Coding System, the average nominal tariff on automobiles is 30 percent, while all the other tariff rates are around or below 20 percent.8 The Malaysia 209 overall tariff structure cascades: the tariffs on final goods are generally higher than the tariffs on production inputs (intermediate and capital goods) (Athukorala 2005). As part of its World Trade Organization commitments, Malaysia has bound 65 percent of its tariff lines. The bound rates are much higher than the applied most favored nation rates (Athukorala 2002).9 This feature of the tariff structure has given the government scope to raise applied tariffs (in 1998), thereby impart- ing a degree of uncertainty to these tariffs. There are no import quotas in Malaysia, and import prohibitions are limited to the prohibitions implemented for national security reasons. By the mid-1990s, only 4.5 percent of all tariff lines involved tariffs that were not ad valorem. This share had declined to 0.7 percent by 2002 because of the additional rationalization of the tariff structure following the Uruguay Round agreement in 1995. There are no tariff quotas or variable import levies (Athukorala 2002). By 2000, the coverage ratio (unweighted) of nontariff barriers in import trade amounted to 2.3 percent, down from the 3.7 percent of the mid-1990s.10 Despite recent tariff increases, the average tariff rate is relatively low by regional standards in terms of both the simple average and the import- weighted average. However, measured by the coefficient of variation, the degree of dispersion among tariff rates is relatively high because of high tariffs for a few product lines, especially motor vehicles (Athukorala 2005). Agricultural trade policy: plantation crops Duties on the two major primary export commodities, rubber and palm oil, were a major source of government revenue until the mid-1980s. Subsequently, duty rates were adjusted in line with world price trends to maintain stable producer prices. Export duties were reduced sharply beginning in the mid-1980s when the viability of some industries was under severe strain because of labor shortages and rising wages, which were propelled by dramatic structural changes in the economy that were driven by export-led industrialization (Ariff and Semudram 1990). The reduction of export duties was aided by tax buoyancy in a rapidly growing econ- omy and by increasing government revenue from petroleum exports. The import duty rates on rubber and palm oil, which increased steadily in the 1960s and 1970s, have declined over the past two decades. In 2000­04, the average annual duty rate was 4.7 percent on rubber and 1.1 percent on palm oil. The higher duty rate on rubber was generated by the additional duty (cess) that con- tinued to be levied to finance the rubber replanting scheme. Duties ranging from 5 to 10 percent are levied on specific grades of crude palm oil to promote domes- tic processing. By 2000, only a few other primary products, such as selected forest products and crude oil, were subject to export duties. Export duties contributed a mere 2 percent to total government revenue (Ministry of Finance 2006). 210 Distortions to Agricultural Incentives in Asia Conventional trade policy and direct government support through funding for research and replanting schemes by and large lost their relevance when the structural changes in the economy began severely to impede the long-term via- bility of plantation crops. Consequently, in recent years, the focus of policy has shifted toward new issues such as forging links between the agricultural sector and the rapidly growing manufacturing sectors, improving the productivity and efficiency of certain agricultural subsectors, and assisting plantation enterprises in relocating to other countries where factor market conditions enable prof- itable production. Relaxing restrictions on labor imports both formally and informally (that is, by turning a blind eye to illicit immigration) has also become an important short-tem measure for reducing labor market pressures (Athukorala 2006a). Agricultural trade policy: food crops Rice is the single most highly assisted crop since the guaranteed price scheme was introduced by the colonial government in 1949 (Ness 1967; Meerman 1979; Pletcher 1989; Zubaidi 1992; Rudner 1994). The emphasis on assistance for paddy farmers gained impetus after independence, particularly as part of the New Economic Policy. The government has assisted rice producers through an all-encompassing guaranteed minimum price scheme, a price subsidy scheme, and a fertilizer sub- sidy. In 1998 (the latest year for which data are available), the total government expenditure on the three schemes amounted to RM 547 million (US$150 million) or about 3 percent of the total value added for this crop. The guaranteed mini- mum price scheme was first introduced in 1949, and the minimum price was sub- sequently adjusted in 1973, 1974, 1979, 1980, 1984, and 1997. Under the scheme, Bernas (a government trading company) undertakes to buy paddy from farmers at no less than the guaranteed minimum price (since 2000, RM 5.49 per kilo- gram). Bernas procures paddy from farmers and mills rice as a business operation. It competes with private millers in the procurement of paddy and the marketing of milled rice. It purchases about 45 percent of the marketable surplus paddy available. Only Bernas is permitted to import rice (at zero duty) into the country. It undertakes to import rice and implement the rice price subsidy program under a long-term contract with the government. The import volume is determined by Bernas according to the shortfall in production over consumption. Rice millers are required to produce 30 percent of their output at standard and premium qual- ity. Bernas is free to determine the price for its superior quality rice, and the prof- its realized on this rice are used to cross-subsidize the minimum production required in standard- and medium-quality rice. Malaysia 211 A cash subsidy for every ton of paddy sold was introduced in 1980 and was increased in 1984 and 1990. Under this scheme, the government makes a fixed payment to farmers (currently RM 2.48 per kilogram) for the paddy sold by them to any commercial rice mills. The subsidy is provided in addition to the guaran- teed minimum price received by farmers. The fertilizer subsidy scheme has been in operation since 1985. There was also a subsidy credit scheme for paddy farmers, but this was terminated in 1996. In 2004, the total government outlay on the price subsidy and the fertilizer subsidy was RM 477 million (about 2 percent of the value added in the paddy subsector). The government also assists paddy farmers by providing drainage and irrigation facilities and management and extension services. The total outlay on these support measures accounted for around 1.5 percent of the total value added in the paddy subsector in 2004 (WTO 2006). Accompanying policies Despite instances of policy slippage, government macroeconomic policy was gen- erally sound over the period under analysis. It supported growth and structural transition in the real sectors of the economy. Budget deficits were mostly kept within prudent limits, while the use of borrowed funds was minimized. The gov- ernment continued to adhere strictly to the colonial policy of avoiding loans from the central bank for budgetary purposes. Overall deficits arose occasionally, but they were financed through noninflationary domestic sources, particularly the private savings accumulated in the Employee's Provident Fund. Moreover, the broadening of the tax base when the economy boomed, coupled with greater effi- ciency in tax collection, brought about a rapid increase in government revenues. For the first time in the history of the country, the government achieved a bal- anced budget in 1993, and budget balance was maintained in subsequent years. Relative to 1986­89, the public sector was a net saver in 1990­96. Fiscal balance as a share of GDP shifted from an annual average deficit of 2.5 percent to a surplus of 1.5 percent between these two periods. Deficit financing reemerged, however, as part of the policy response to the 1997­98 financial crisis. The budget deficit as a share of GDP rose from 1.8 to 5.2 percent in 1998­2002. The share fell thereafter, reaching 3.5 percent in 2006. By developing countrywide standards, Bank Negara Malaysia (the central bank) has maintained an impressive track record in maintaining domestic price stability and averting real exchange rate misalignment (Corden 1996; Athukorala 2001). As part of the macroeconomic adjustment package launched in 1986, greater flexibility was introduced to the basket peg. The policy of Bank Negara Malaysia involved allowing the exchange rate to reflect underlying trends in the economy, while intervening in the foreign exchange market to smooth the 212 Distortions to Agricultural Incentives in Asia excessive fluctuation in exchange rate movements that was caused by fluctuations in short-term capital inflows. This policy was successful in achieving significant depreciation in the overall real exchange rate from 1987 to about 1993. There was a mild appreciation of the real exchange rate in the three years leading up to the financial crisis in 1997. This reflected macroeconomic imbalances in the booming economy. The crisis was instrumental in reversing the appreciation. By 2000, although the exchange rate of the ringgit had been fixed against the U.S. dollar (at RM 3.8 per dollar) in September 1998 and despite the deficit financing that had been undertaken as part of the crisis management package, the real exchange rate had depreciated by almost 20 percent against precrisis levels.11 It has remained virtually unchanged since then. Athukorala and Loke (2007) show that, over the past three decades, the real exchange rate in the export crop subsector has been behaving quite differently from the overall real exchange rate. Notwithstanding periodic depreciations trig- gered by increases in world commodity prices, the exchange rate in the export crop subsector has appreciated steadily over the past 25 years. This contrast reflects the ongoing structural transformation of the economy that has resulted in a deteriora- tion in the relative profitability of the traditional plantation crop subsector. Trends and Patterns in Distortions in Agricultural Incentives In this section, we provide an analysis of the changing scope of and patterns in direct and indirect distortions in the incentives faced by domestic agriculture in Malaysia. We use the methodology developed by Anderson et al. (2008) and described in appendix A. The main focus of the methodology in our study is government-imposed distortions that create a gap between domestic prices and prices as they would be under free-market conditions. Since it is not possible to understand the characteristics of agricultural development through a sectoral view alone, we estimate not only the effects of direct agricultural policy measures, but also, for comparative evaluation, the distortions in nonagricultural tradables. Specifically, we compare the nominal rates of assistance (NRAs) for tradable farm products with the NRAs for nonagricultural tradables by calculating indicators of the relative rate of assistance (RRA). In our calculations, we assume that the agri- cultural products that we do not cover have an average NRA of zero, and we assume that the shares of noncovered farm production are one-third each for exportables, importables, and nontradables. We have been unable to avoid two important limitations in our estimates because of a lack of data. First, in the case of the three plantation (cash) crops, rubber, palm oil, and cocoa, we ignore the potential differences between border Malaysia 213 (reference) prices and domestic prices that arise because of quality differences. This may infuse an underestimation bias into our calculations. Second, in all cases, we have assumed that there is a complete pass-through to farmgate prices of any change in wholesale prices. This may introduce an upward bias into our esti- mates. These limitations are only important, however, in comparisons among the effects on incentives among products or across countries at a given moment. They are unlikely to distort inferences based on intertemporal comparisons (changes in incentives over time) because the magnitude of the bias is less liable to vary over time. It is also important to note that, because of our estimation method, our RRA estimates do not fully capture the distortions in agricultural incentives arising from changes in the tariffs on tradable inputs. Given the cascading nature of Malaysia's tariff structure, this is a potentially important source of downward bias in the RRA estimates (Athukorala 2006b). Our estimates of the direct distortions in the incentives for covered products are reported as five-year averages in table 5.3, and they are shown as annual aver- ages in summary form in figure 5.2. The average NRAs for all covered products were negative from 1960 to 1984, but the magnitude of the NRAs declined during the period. The five-year average fluctuated between 0 and 3 percent starting in the mid-1980s. However, this aggregate picture conceals the significant assistance provided to paddy farmers. NRA estimates for individual commodities point to broadly similar patterns in the changes in the incentives faced by the two most important plantation prod- ucts, rubber and palm oil (table 5.3). In both cases, the NRAs were negative throughout, but the absolute magnitude declined sharply over the two most recent decades. This reflected cuts in export duties. However, with the exception of some of the earlier years, the negative incentives in the palm oil industry were much lower in magnitude relative to the incentives in the rubber industry. For the entire period of 1960­2004, the annual average NRA for palm oil was 7.5, com- pared with 11.5 for rubber. Given that the fortunes of both products have been predominantly determined by domestic resource-pull effects that have arisen because of rapid structural adjustment in the wider economy, the generally high negative assistance to rubber relative to palm oil remains a puzzling feature of the structure of the incentives. Cocoa was never taxed heavily because it was always considered a minor export crop. The NRA for this product varied between 0 and 3 percent over the period. Among the four products under study, paddy rice is notable for a persistently high rate of assistance. The average NRA for paddy and rice in the 1960s and early 1970s was 8.5 percent, although there was a high degree of annual fluctuation. The average rice NRA was nearly 40 percent in 1975­79 following an upward adjustment in the guaranteed minimum price. It then reached a peak average of 214 Table 5.3. NRAs for Covered Agricultural Products, Malaysia, 1960­2004 (percent) Indicator, product 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Exportablesa 12.1 9.6 13.4 20.0 12.8 5.6 4.7 3.6 1.6 Palm oil 11.4 10.6 15.2 15.0 5.8 3.2 3.1 3.0 1.1 Cocoa 0.0 1.2 2.8 1.7 1.5 1.4 2.3 2.1 0.0 Rubber 12.1 9.5 12.8 22.5 18.2 8.7 8.1 6.8 4.7 Import-competing productsa 19.1 1.9 3.1 39.2 93.8 158.0 127.2 57.4 71.0 Rice 19.1 1.9 3.1 39.2 93.8 158.0 127.2 57.4 71.0 All covered productsa 8.4 8.7 10.5 15.3 5.7 1.8 3.4 0.3 2.4 Dispersion, covered productsb 30.6 18.5 21.1 43.8 53.4 65.8 57.3 36.7 43.2 Coverage, at undistorted prices 86 86 86 85 80 75 67 59 57 Source: Athukorala and Loke 2007. a. Weighted averages; the weights are based on the unassisted value of production. b. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of covered products. Malaysia 215 Figure 5.2. NRAs for Exportable, Import-Competing, and All Agricultural Products, Malaysia, 1960­2004 300 250 200 % 150 NRA, 100 50 0 50 19601962196419661968197019721974197619781980198219841986198819901992199419961998200020022004 year import-competing products total exportables Source: Athukorala and Loke 2007. 158 percent during the five years from 1985 to 1989 following the introduction of a price subsidy (over and above the guaranteed minimum price). But, during the two decades following the macroeconomic crisis in 1985­87, the NRAs for paddy more than halved, although, in 2000­04, it was still above 70 percent. The disag- gregated data show that the farmgate price of paddy continued to be high and that there were only periodic upward shifts resulting from increases in the guaranteed minimum price and the price subsidy (Athukorala and Loke 2007). In this con- text, the year-to-year variations in the NRAs arose mostly because of changes in the reference (border) price. For instance, the dramatic decline in NRAs from 127 percent in 1990­94 to 57 percent in 1995­99 was brought about by a sharp decline in world rice prices between these two periods. The NRAs then increased to 71 percent, which reflected the recovery in world prices. Finally, a comparison of the weighted average NRAs for the exportables (rubber, palm oil, and cocoa) and the importables (which, in our case, is limited solely to paddy) points to a persistent bias in agricultural incentives in favor of import- competing production relative to export-oriented production (see the trade bias index shown in table 5.4). Based on similar estimates for 1960­82, Jenkins and Lai (1991) inferred that the excessive protection accorded to paddy farmers had a neg- ative effect on the expansion of export agriculture. This inference does not seem 216 Table 5.4. NRAs in Agriculture Relative to Nonagricultural Industries, Malaysia, 1960­2004 (percent) Indicator 1960­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Covered productsa 8.4 8.7 10.5 15.3 5.7 1.8 3.4 0.3 2.4 Noncovered products 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 All agricultural productsa 7.2 7.5 9.0 13.0 4.6 1.3 2.3 0.2 1.3 Non-product-specific assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total agricultural NRAb 7.2 7.5 9.0 13.0 4.6 1.3 2.3 0.2 1.3 Trade bias indexc 0.22 0.06 0.14 0.31 0.35 0.33 0.28 0.12 0.12 NRA, all agricultural tradables 7.6 7.9 9.4 13.7 4.9 1.4 2.6 0.2 1.5 NRA, all nonagricultural tradables 7.4 7.0 7.1 6.5 5.2 3.9 2.8 2.0 0.9 RRAd 14.0 13.9 15.5 18.9 9.6 2.4 0.3 2.2 0.6 Source: Athukorala and Loke 2007. a. Including product-specific input subsidies. b. Including product-specific input subsidies and non-product-specific assistance. The total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. c. The trade bias index is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. d. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. Malaysia 217 Figure 5.3. NRAs for Agricultural and Nonagricultural Tradables and the RRA, Malaysia, 1960­2004 15 10 5 0 cent 5 per 10 15 20 25 19601962196419661968197019721974197619781980198219841986198819901992199419961998200020022004 year NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Athukorala and Loke 2007. Note: For the calculation of the RRA, see table 5.4, note d. valid for the period starting around the late 1980s. The steady deterioration in the profitability of export-oriented agriculture, as well as paddy production, was rooted mainly in the ongoing process of structural transformation in the wider economy. Nonetheless, the significant assistance for paddy producers was presum- ably a major source of the distortions within the food-crop subsector. These distor- tions constrained resource reallocation from the structurally weak paddy subsector to high-value food production for the domestic and export markets. The NRAs for nonagricultural tradables, which recorded a mild decline in the 1960s and 1970s, plummeted thereafter and had reached almost zero by around 2000 (figure 5.3). Direct tariff cuts and the rapid expansion in export-oriented manufacturing, which enjoys duty-free status for all imported inputs in the pro- duction process, contributed to this decline. Disaggregated data (for brevity, not reported here) show that the rapid expansion in export-oriented manufacturing continued to act as a more powerful force than the direct tariff cuts. As a consequence of these changes in agricultural and nonagricultural assis- tance, average RRAs gradually shifted from larger negative levels in the earlier decades to almost zero (table 5.4). However, this does not mean that there would be no economic gains from additional policy reform. Indeed, as shown at the bot- tom of table 5.3, the dispersion of NRAs within the farm sector has not declined 218 Distortions to Agricultural Incentives in Asia much over time; so, there is still scope for improved resource reallocation if the assistance for paddy production is phased out. Concluding Remarks Malaysia stands out among developing countries because of the government's long-standing commitment to the maintenance of a relatively open trade and investment policy regime. The government has persistently avoided heavy reliance on quantitative restrictions and other forms of nontariff protection. Tariffs on domestic manufacturing and agriculture continue to be low relative to the corre- sponding tariffs in other developing countries. Export taxes, which were impor- tant sources of government revenue until about the mid-1980s, were reduced when the plantation subsector experienced severe cost pressure because of rapid growth and structural change under export-led industrialization. The average level of import tariffs also declined significantly, notwithstanding the periodic upward adjustment in some tariffs and the special case of the heavy protection in the automotive industry. The government's record of commitment to openness is particularly remarkable in that it reflects unilateral and voluntary policy choices rather than pressure from major trading partners or from conditions imposed by multilateral donor agencies or resulting from negotiations under the auspices of the World Trade Organization. Nonetheless, there are notable anomalies in the incentive structure in Malaysia that encourage the channeling of resources into inefficient activities. In particular, the tariff structure is characterized by a dualistic pattern whereby export-oriented production occurs within a virtual free trade regime, side-by-side with produc- tion predominantly oriented toward the domestic market in manufacturing and agriculture and assisted through tariff protection. The tariff rate structure is also characterized by a high degree of dispersion because of high tariff peaks on a few product lines and growing reliance on nonautomatic import licensing to regulate the imports of a significant number of products that directly compete with the domestic production carried out by public sector enterprises. This substantial departure from neutrality implies that there is ample room for policy discretion, as opposed to pure economic policy, in the effort to influence resource allocation in the economy. The excessive assistance for paddy farmers remains a major distortion in agri- cultural incentives. In addition to the obvious welfare implications, this anomaly presumably hinders the diversification of domestic agriculture toward new dynamic product lines. Given the ongoing process of dramatic structural trans- formation in the economy that has ushered in an era of massive urban-to-rural labor migration and generates cost pressures on traditional agriculture, the case Malaysia 219 for protecting paddy farmers on self-sufficiency grounds has lost relevance. The outright dismantling of assistance is a nonoption because of political economy considerations. Nonetheless, there is a strong case for replacing the existing com- plicated and costly incentives with direct income support for farmers. The fiscal burden of this support is unlikely to be high because the agricultural labor force has been rapidly depleting, and the incidence of rural poverty, though relatively high by national standards, has been declining. This issue deserves additional sys- tematic analysis. Notes 1. By 1984, the Federal Land Development Authority accounted for 28 percent of the 1.3 million hectares under oil palms. 2. The bulk of the land classified as smallholding in these data involves farmers who participate in large government-run plantation schemes. Smallholders who do not participate in these schemes account for only 8 percent of the planted area measured in hectares. 3. Powerful forces on both the demand side and the supply side have underpinned this structural shift. On the demand side, the internationalization of food habits--the growing significance of imported processed items in consumption patterns in developed countries, as well as among large sec- tions of populations in many developing countries--appears to have played a key role. Factors such as international migration, the communications revolution, and tourism have contributed to the phenom- enon. This significant demand-side impetus seems to have been supported by important supply-side developments such as improvements in food technology, refrigeration facilities, and transportation that have fostered international trade in various processed food products that are generally highly perishable. Indonesia is more well placed than Malaysia to benefit from this structural shift in world food trade given its rich agricultural resource base and greater availability of labor (food processing and food packaging for export are highly labor intensive). 4. The Federation of Malaya, comprising 11 states in the Malay Peninsula, secured independence from Britain on August 31, 1957. Sabah, Sarawak, and Singapore joined Malaya to form Malaysia on September 16, 1963. Singapore left the federation in August 1965. 5. The emergence of three identifiable and mutually exclusive ethnic groups as distinct, self- conscious groups (Chinese, Indian, and Malay) stemmed in substantial part from the needs and prior- ities of British colonial policy (Ganguly 2003). 6. In the Rosentein-Rodan (1961) growth trajectory up to 1976 for 66 countries considered devel- oping countries at the time, Malaysia was classified in the low-growth category, together with Indonesia; the Republic of Korea; Singapore; Taiwan, China; and Thailand. 7. In a recent comprehensive study of the patterns and chronology of trade policy reforms, Sachs and Warner (1995) identify Malaysia as one of eight developing countries in which trade regimes remained open throughout the post­World War II period. 8. Unless otherwise noted, we have taken the tariff rates reported in this chapter from the latest (2003) tariff schedule available in the APEC Tariff Database. 9. In 2002, the simple average bound, unbound, and applied tariff rates were 19, 35, and 9.2 per- cent, respectively. All agricultural tariff lines were bound, but at much higher average levels than the manufacturing tariff lines. 10. Calculated as a percent share in total imports of the import value of the Harmonized Com- modity Description and Coding System 6 tariff lines affected by nontariff barriers. 11. The ringgit peg to the U.S. dollar was eventually abandoned in favor of a managed float system. The system went into effect on July 21, 2005. 220 Distortions to Agricultural Incentives in Asia References Ahmad, T. M. A. T., and A. Tawang. 1999. "Effects of Trade Liberalization on Agriculture in Malaysia: Commodity Aspects." CGPRT Centre Working Paper 46, Regional Co-ordination Center for Research and Development of Coarse Grains, Pulses, Roots, and Tuber Crops in the Humid Trop- ics of Asia and the Pacific, United Nations Economic and Social Commission for Asia and the Pacific, Bogor, Indonesia. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. APEC Tariff Database (Asia Pacific Economic Cooperation Tariff Database). Asia-Pacific Economic Cooperation. http://www.apectariff.org/ (accessed May 2008). Ariff, M., and M. Semudram. 1990. "Malaysia." In Trade, Finance, and Developing Countries: Strategies and Constraints in the 1990s, ed. S. Page, 23­55. London: Harvester Wheatsheaf. Athukorala, P.-C. 1991. "An Analysis of Demand and Supply Factors in Agricultural Exports from Developing Asian Countries." Weltwirtschaftliches Archiv 127 (4): 764­91. ------. 2001. Crisis and Recovery in Malaysia: The Role of Capital Controls. Cheltenham, UK: Edward Elgar. ------. 2002. "Malaysian Trade Policy and the WTO Trade Policy Review 2001." World Economy 25 (9): 1299­1317. ------. 2005. "Trade Policy in Malaysia: Liberalization Process, Structure of Protection, and Reform Agenda." ASEAN Economic Bulletin 22 (1): 19­34. ------. 2006a. "International Labour Migration in East Asia: Trends, Patterns, and Policy Issues." Asian-Pacific Economic Literature 20 (1): 18­39. ------. 2006b. "Post-Crisis Export Performance: The Indonesian Experience in Regional Perspective." Bulletin of Indonesian Economic Studies 42 (2): 177­211. Athukorala, P.-C., and S. K. Jayasuriya. 2003. "Food Safety Issues, Trade, and WTO Rules: A Developing Country Perspective." World Economy 26 (9): 141­62. Athukorala, P.-C., and W.-H. Loke. 2007. "Distortions to Agricultural Incentives in Malaysia." Agricul- tural Distortions Working Paper 27, World Bank, Washington, DC. Athukorala, P.-C., and C. Manning. 1999. Structural Change and International Migration in East Asia: Adjusting to Labour Scarcity. Melbourne: Oxford University Press. Athukorala, P.-C., and J. Menon. 1999. "Outward Orientation and Economic Performance: The Malaysian Experience." World Economy 22 (8): 1119­39. Bank Negara Malaysia. various issues. Monthly Bulletin of Statistics. Kuala Lumpur: Bank Negara Malaysia. Barlow, C. 1978. The Natural Rubber Industry: Its Development, Technology, and Economy in Malaysia. Kuala Lumpur: Oxford University Press. ------. 1997. "Growth, Structural Change, and Plantation Tree Crops: The Case of Rubber." World Development 25 (10): 1589­1607. Barlow, C., and S. Jayasuriya. 1987. "Structural Change and Its Impact on Traditional Agricultural Sectors of Rapidly Developing Countries: The Case of Natural Rubber." Agricultural Economics 1 (2): 159­74. Corden, W. M. 1996. Pragmatic Orthodoxy: Macroeconomic Policies in Seven East Asian Economies. San Francisco: International Center for Economic Growth. Economic Planning Unit. 2001. Eighth Malaysia Plan. Putrajaya, Malaysia: Economic Planning Unit, Prime Minister's Department. ------. 2006. Ninth Malaysia Plan. Putrajaya, Malaysia: Economic Planning Unit, Prime Minister's Department. Ganguly, S. 2003. "Ethnic Policies and Political Quiescence in Malaysia and Singapore." In Government Policies and Ethnic Relations in Asia and the Pacific, ed. M. E. Brown and S. Ganguly, 233­72. Cambridge, MA: MIT Press. Malaysia 221 Government of Malaysia. 1971. Second Malaysia Plan, 1971­1975. Kuala Lumpur: Government Printers. Jenkins, G. P., and A. Lai. 1989. Trade, Exchange Rate, and Agricultural Policies in Malaysia. World Bank Comparative Study. Washington, DC: World Bank. ------. 1991. "Malaysia." In Asia. Vol. 2 of The Political Economy of Agricultural Pricing Policy, ed. A. O. Krueger, M. Schiff, and A. Valdés, 67­106. World Bank Comparative Study. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Khoo, K. M., and D. Chandramohan. 2002. "Malaysian Palm Oil Industry at Crossroads and Its Future Direction. Oil Palm Industry Economic Journal 2 (2): 10­15. Levin, J. V. 1960. The Export Economies. New Haven, CT: Yale University Press. Lim, D. 1992. "The Dynamics of Economic Policy-Making: A Study of Malaysian Trade Policies and Performance." In The Dynamics of Economic Policy Reforms in South-East Asia and the South-West Pacific, ed. A. J. MacIntyre and K. Jayasuriya, 94­114. Singapore: Oxford University Press. Lin, T. Y. 1984. "Inter-ethnic Restructuring in Malaysia, 1970­80: The Employment Perspective." In From Independence to Statehood: Managing Ethnic Conflict in Five African and Asian States, ed. R. B. Goldman and J. Jayaratnam Wilson, 44­61. London: Frances Printer. MacBean, A. I. 1989. "Agricultural Exports of Developing Countries: Market Conditions and National Policies." In The Balance between Industry and Agriculture in Economic Development, ed. N. Islam, 129­64. London: Macmillan. Meerman, J. 1979. Public Expenditure in Malaysia: Who Benefits and Why? New York: Oxford Univer- sity Press. Ministry of Agriculture. 1997. Third National Agricultural Policy, 1998­2010. Kuala Lumpur: Ministry of Agriculture. Ministry of Finance. 2005. Economic Report 2005/2006. Kuala Lumpur: Ministry of Finance. ------. 2006. Economic Report 2006/2007. Kuala Lumpur: Ministry of Finance. ------. various issues. Economic Report. Kuala Lumpur: Ministry of Finance. Myint, H. 1984. "Inward and Outward-Looking Countries Revisited: The Case of Indonesia." Bulletin of Indonesian Economic Studies 20 (2): 39­52. Ness, G. D. 1967. Bureaucracy and Rural Development in Malaysia. Berkeley, CA: University of California Press. Pletcher, J. 1989. "Rice and Padi Market Management in West Malaysia, 1957­86." Journal of Develop- ing Areas 23 (April): 363­84. ------. 1991. "Regulation with Growth: The Political Economy of Palm Oil in Malaysia." World Devel- opment 19 (6): 623­36. Power, J. H. 1971. "Structure of Protection in West Malaysia." In The Structure of Protection in Develop- ing Countries, ed. B. Balassa, 203­22. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Rosentein-Rodan, P. 1961. "International Aid for Underdeveloped Countries." Review of Economics and Statistics 43 (2): 107­38. Rudner, M. 1994. Malaysian Development: A Retrospective. Ottawa: Carleton University Press. Sachs, J. D., and A. Warner. 1995. "Economic Reforms and the Process of Global Integration." Brookings Papers on Economic Activity 26 (1): 1­118. Snodgrass, D. R. 1980. Inequality and Economic Development in Malaysia. Kuala Lumpur: Oxford University Press. World Bank. 1955. The Economic Development of Malaya. Baltimore: Johns Hopkins University Press. WTO (World Trade Organization). 2006. Trade Policy Review: Malaysia. Geneva: WTO. Zubaidi, A. 1992. "The Welfare Cost of Malaysian Rice Policy under Alternative Regimes." Malaysian Journal of Economic Studies 29 (2): 1­12. 6 THE PHILIPPINES Cristina David, Ponciano Intal, and Arsenio M. Balisacan The economic performance of the Philippines has lagged behind that of most other developing countries in Asia. Whereas the Philippine economy and agricul- tural sector performed moderately well in the 1960s and 1970s because of the early advent of the Green Revolution in rice and the world commodity boom, the country has shown the lowest average growth rates in South and Southeast Asia in gross domestic product (GDP), gross value added (GVA) in agriculture, and agri- cultural exports over the past two decades (table 6.1). Previous studies have argued that the country's poor agricultural performance has been caused largely by weaknesses in the policies and institutional framework governing the sector and less by real domestic and external market factors (David 2003; Balisacan, Fuwa, and Debuque 2004). Government price and trade policies have distorted economic incentives, and the choice of policy instruments has pro- moted rent seeking and raised the economic cost of government intervention. The lack of market infrastructure, underinvestment in agricultural research, distor- tions in land markets because of the agrarian reform program, and other weak- nesses in governance have all contributed to the poor performance in the sector. The declining trend in tariff protection since the 1980s through a series of uni- lateral tariff reform programs and multilateral and regional trade agreements are well documented (Manasan and Pineda 1999; Aldaba 2005a, 2005b; Pasadilla 2006). Ex ante assessments of the welfare impacts of trade liberalization that rely on computable general equilibrium models consistently report positive effects (Clarete 1991; Cororaton 2000; Habito and Cororaton 2000). International The authors are grateful to Ernesto Valenzuela for computational assistance. 223 224 Table 6.1. Agriculture in the Economy, Growth Rates, the Philippines and Other Asian Countries, 1960­2004 (percent) 1960­80 1980­2004 Country GDP Agriculture, GVA Agriculture, exports GDP Agriculture, GVA Agriculture, exports Philippines 5.3 4.1 12.0 2.7 2.0 1.6 Bangladesh 2.6 1.6 3.1 4.3 2.7 1.3 China 5.5 4.3 15.0 9.5 4.2 6.6 India 3.6 2.2 9.0 5.4 3.4 5.8 Indonesiaa 7.9 4.6 11.6 5.4 2.9 7.2 Malaysiab 7.2 4.8 12.7 6.6 2.2 5.7 Pakistan 5.8 3.6 14.2 4.7 3.7 7.2 Thailand 7.5 4.8 13.5 6.0 2.7 6.7 Vietnamc -- -- -- 6.6 3.7 18.2 Source: Author compilation based on data in World Development Indicators Database 2008; FAOSTAT Database 2008. Note: Growth rates have been estimated through the regression method. -- no data are available. a. 1960­80: GDP and GVA data refer to 1970­80. b. 1960­80: GVA data refer to 1970­80. c. 1980­2004: GVA data refer to 1985­2004. The Philippines 225 Table 6.2. Changes in Agricultural Structure and Trade Openness, the Philippines, 1960­2004 (percent) Indicator 1960 1970 1980 1990 2000 2004 Share of agriculture GDP 30 28 24 22 20 14 Employment 61 52 48 45 37 37 Importsa 19 14 8 10 9 8 Exports 64 44 35 15 5 6 Indicators of trade openness Share of agricultural imports 6 [9] 10 [16] 9 [23] 13 [27] 22 [45] 26 [51] in the GVAa Share of agricultural exports 33 [9] 44 [15] 26 [17] 14 [19] 15 [51] 19 [46] in the GVA Total agricultural imports and 38 [18] 54 [36] 35 [43] 28 [48] 37 [96] 45 [97] exports in the GVA Sources: Compiled based on data of the National Statistical Coordination Board; the Bureau of Labor and Employment Statistics, Department of Labor and Employment; and the National Statistics Office. Note: The numbers in square brackets refer to the corresponding indicators of trade openness in the entire economy. a. For 2004: if agricultural inputs are included, the ratio of agricultural imports to the GVA in agriculture would be significantly higher, reaching 39 percent; the ratio of agricultural inputs to total imports based on this broader definition would be 11 percent. studies, mostly using the Global Trade Analysis Project model, likewise generally show that the results of trade reform are favorable. Yet, after more than two decades of effort at trade liberalization, the country has not enhanced its eco- nomic performance significantly. Per capita income continues to stagnate; domes- tic employment opportunities remain limited; and poverty reduction lag relative to the progress being accomplished in most of the country's Asian neighbors. Although the economy appears to have become more open if one considers the substantial rise in the ratio of the traded value of imports and exports to GDP (table 6.2), the rise in this ratio has not been accompanied by growth in GDP. The improvement in the indicators of trade openness has been achieved primarily through rapid export growth in semiconductors and electronic components that exhibit high shares of imported content and low shares of value added. The growth rate in food and agricultural exports continue to drop, and the country's dependence on agricultural imports has risen sharply (table 6.2).1 Several complex factors may help explain why the predicted impacts of trade liberalization have not been realized. One explanation, at least with respect to the agricultural sector, may be that the rate of trade liberalization measured according to the trends in average tariffs (typically used as indicators by local and 226 Distortions to Agricultural Incentives in Asia international analysts) does not accurately reflect the extent or the direction of change in agricultural protection during recent decades. The objective of our study is to quantify trends and patterns in agricultural dis- tortions from the early 1960s to 2004 and explain changes in these trends and pat- terns over time. In the next section, historical patterns in agricultural performance and structural change are briefly described. The subsequent section examines the evolution of economy-wide and agriculture-specific policies that have distorted price incentives in the sector. In the following section, we discuss the estimated impact of these policies on agricultural incentives. The penultimate section ana- lyzes the reasons behind policy choices, including the role of multilateral and regional trade agreements in the changes observed in recent years. Finally, we draw out the prospects for national policy reform, including the likely versus the desir- able policy direction through to 2020, the implications for the choice of policy instruments and for the trend level in the distortions in agricultural incentives, and the policy lessons for other developing and transition economies. Agricultural Performance and Structural Change Despite the relatively slow growth of the Philippine economy, the structural trans- formations that have occurred in the course of economic development have not been unusual. The decline in the contribution of agriculture to GDP was rather slow from 1960 to 1980 (from 30 to 24 percent), but has become more rapid in recent years (14 percent by 2004). The decline was accompanied by a steady drop in the share of the sector in total employment from 61 to 37 percent from 1960 to 2000. The share has remained constant since then (table 6.2). Unlike the rapid industrialization that characterized economic growth among the Asian tigers, however, the share of industry in the Philippines rose from 31 to 39 percent in 1960­80, but then fell, reaching around 30 percent by 2004. Food manufacturing accounts for about 40 percent of the industrial sector. Half of this share consists of light processing of rice, maize, sugar, coconuts, livestock, and poultry. Services have accounted for the largest contribution to GDP and total employment. While the growth in services has been driven primarily by domestic demand for logistics, trading, and financial services, the rapid expansion in busi- ness process outsourcing has been changing the nature and the prospects for growth of the sector in recent years. Growth rate and composition Average annual growth rates in the GVA of major agricultural commodities have been quite erratic. The crop subsector grew rapidly prior to 1980 because of the The Philippines 227 Table 6.3. GVA Growth Rates, Major Agricultural Commodities, the Philippines, 1960­2004 (at constant 1985 prices, percent) Product 1960­70 1970­80 1980­90 1990­2004 Crops 4.3 6.2 1.1 1.9 Rice -- 4.3 2.9 3.6 Maize -- 5.2 3.1 1.9 Coconuts, including copra -- 7.8 3.9 1.1 Sugar -- 5.2 1.8 4.1 Bananas -- 13.9 1.8 4.0 Other crops -- 8.1 2.2 0.8 Livestock and poultry 3.1 3.1 5.7 4.4 Livestock -- 0.8 4.8 3.5 Poultry -- 8.5 7.5 5.7 Source: Compiled based on data of the National Statistical Coordination Board. Note: -- no data are available. Green Revolution in rice production and the world commodity boom, but it per- formed poorly thereafter, when the average growth rate was far below the growth rate of the population. The general slowdown may be observed across commodi- ties (table 6.3). Rice is the main staple and the single most important crop. While the growth rate in rice production declined beginning in 1970, it was higher than the popula- tion growth rate. Rice continues to receive the bulk of public expenditure in the crop subsector, and it has also benefited from increasing price protection (see below). Rice imports as a share of the total supply of rice have risen since the 1980s, reflecting the effect of rising incomes and a shift away from maize as a food staple. Maize production experienced declining growth rates despite rising price protection and a rapid expansion in the demand for maize as feed in the pig meat and poultry industries. The poorest performers are the traditional export crops such as coconuts, sugar, abaca, and tobacco, each of which showed a decreasing growth rate. In con- trast, nontraditional export crops such as bananas, pineapples, and mangoes showed high growth rates. However, crop diversification, particularly toward high-value horticultural crops that raised the growth rates of the agricultural sec- tors in Chile, Thailand, and other developing countries, was not as pronounced in the case of the Philippines. The growth rate in livestock production accelerated after 1980. The contribu- tion of livestock to the GVA rose from 18 to nearly 25 percent within only 25 years. This remarkable performance was generated by increasing domestic 228 Distortions to Agricultural Incentives in Asia demand, as well as productivity gains resulting from the shift to larger-scale oper- ations and the adoption of the new technologies embedded in imported breeds, veterinary medicines, and feed ingredients. Agricultural trade and trade openness Agriculture has historically been a net earner of foreign exchange. In the 1960s, agricultural exports accounted for nearly two-thirds of total exports, while agri- cultural imports accounted for less than 20 percent of total imports (table 6.2). The share of the sector in total exports began dropping sharply in the 1990s. By 2004, the sector had ceased to be a net earner of foreign exchange; the ratio of agricultural imports to agricultural exports rose from about one to three in the 1960s and 1970s to around four to three in 2004. The relatively high growth rate of agricultural exports in the 1970s arose mainly because of the world commodity boom and the expansion of nontraditional commodities (bananas, pineapples, and fishery products). However, world commodity prices fell sharply in the 1980s and have remained low until recently. Meanwhile, the growth in nontraditional agricultural exports had leveled off by the 1990s. In stark contrast, neighboring countries experienced a major export boom in cash crops during the same period. Thus, Thailand showed dramatic success in rubber, Malaysia in palm oil, and Indonesia in palm oil and cocoa. The composition of agricultural exports has been changing in the Philippines. Coconut products has continued to be the top earner of foreign exchange, although their share in agricultural exports has decreased from nearly 70 percent in 1970 to less than 30 percent recently. The contribution of sugar to agricultural exports, which was second only to coconuts in the 1970s (30 percent), is now only 3 percent. The export value of bananas has been about twice that of sugar; and this has been exceeded by the export value of pineapples since the 1990s. Fruits and vegetables now account for nearly 30 percent of agricultural exports.2 The rapid growth in agricultural imports relative to agricultural exports has stemmed from several factors. First, economic development has increased the demand for food products exhibiting higher income elasticities. The Philippines does not have a comparative advantage in many of these products, including wheat, beef, milk, and other dairy products. Second, livestock and poultry require agricultural inputs, such as soybean meal, maize, fishmeal, and other feed ingredi- ents, that are cheaper to import than to produce domestically. Third, agricultural modernization has led to greater reliance on modern manufactured inputs that are mostly imported, such as fertilizers, agricultural chemicals, farm and agropro- cessing machinery, and veterinary medicines. Fourth, trade liberalization has gen- erated more imports of once highly protected agricultural commodities such as fruits and beef. The Philippines 229 If one measures trade openness by imports and exports as a share of the GVA, then there has been a decline in agriculture's trade openness (table 6.2). While agriculture was more open relative to the rest of the economy up to the 1970s, the reverse has been true since then. This shift has been caused not so much by the reduction in the ratio of imports to the GVA (because these have increased in both the agricultural and nonagricultural sectors), but rather by the declining export ratios in agriculture in contrast to the steadily rising trend in the rest of the econ- omy. The declining trend in the trade openness of agriculture in the 1970s and 1980s gradually reversed in the 1990s, but the rate of the rise in import ratios con- tinued to be higher than the rate of the rise in export ratios. Comparative advantage and trends in productivity The growth of agriculture has occurred more slowly in the Philippines than in other developing Asian economies. The stagnation in agricultural exports from the Philippines suggests that the country has been losing its comparative advan- tage in the sector. Indeed, measures of revealed comparative advantage show a sharp drop for Philippine agriculture as a whole, as well as for all major Philippine agricultural exports (table 6.4). For example, the country's share of the world market in coconut products has fallen, and sugar is now also imported. (Exports Table 6.4. Revealed Comparative Advantage, Major Agricultural Commodities, the Philippines, 1960­2004 (percent) Product 1960 1970 1980 1990 2000 2004 All agriculturea 3.0 2.6 2.9 1.6 0.6 0.8 Coconuts 116 145 224 212 71 97 Sugarb 18 21 12 4 1 1 Bananas -- -- 3 23 11 14 Pineapples Canned -- -- 82 70 27 29 Fresh -- -- 49 55 10 8 Source: Compiled based on data in FAOSTAT Database 2008. Note: Comparative advantage is calculated as the ratio of the share of a commodity group in the total exports of a country to the share of the same commodity group in total worldwide exports. Except for 1960 and 2004, the data represent three-year averages centered on the year shown. -- no data are available. a. Including fisheries. b. Sugar has historically been exported to the United States at a preferential price (that is, at higher than world prices). Hence, a value greater than unity in this case does not reveal comparative advantage. The sharp declining trend may nonetheless be interpreted as an indicator of a rapid deterioration in comparative advantage. 230 Distortions to Agricultural Incentives in Asia have been limited to shipments to the high-priced U.S. market where they have benefited from preferential access.) Even among nontraditional exports, such as bananas and pineapples, the shares of Philippine products on world markets have been declining since the mid-1980s. The apparent loss in comparative advantage in agriculture is consistent with trends reported in the indicators of labor and land productivity (David 2003). Labor productivity and land productivity both increased up to the late 1970s, par- ticularly during the Green Revolution in rice production. Whereas labor produc- tivity in agriculture as a whole recovered after a sharp drop in the early 1980s, labor productivity in the crop subsector has stagnated since then. Land productiv- ity in the crop subsector has grown slowly, particularly recently; yields per hectare among traditional exports have generally remained constant or have declined. Higher growth in yields may be observed in rice, maize, and nontraditional exports such as bananas, pineapples, and mangoes. However, productivity growth appears to have occurred in the livestock subsector, where international technol- ogy transfers, the greater scale of operations, and other management-related innovations have increased production efficiencies significantly. The Historical Evolution of Price Intervention Policies It is helpful to describe policy trends since the 1960s. We begin with economy- wide policies and then turn to policies specific to agriculture. Economy-wide policies An import-substitution industrialization strategy dominated Philippine eco- nomic policies up to the late 1970s. The groundwork for the strategy involved comprehensive foreign exchange and import controls that were established in response to the severe balance of payments crisis that occurred in the late 1940s, shortly after the country's political independence from the United States. The gov- ernment's use of essentiality criteria in allocating foreign exchange and import licenses during the 1950s encouraged domestic production at the final stages of the production of primarily nonessential and semiessential consumer goods rather than backward integration in the production of raw materials and interme- diate and capital goods. These policies defended an overvalued peso and thus clearly penalized exports and agriculture. A tariff system was instituted as a decontrol measure in 1957. However, the sys- tem largely preserved the bias in the incentive structure: tariffs depended on essentiality criteria. Import duties were higher for semifinished products than for The Philippines 231 raw materials and capital goods and even higher for finished products. Moreover, quantitative trade restrictions continued to be implemented on a substantial number of agricultural and nonagricultural products. Indeed, the balance of pay- ments problem encountered in the early 1960s rendered tariff protection redun- dant; import and foreign exchange controls were predominant. The adoption of a multiple exchange rate system added to the penalization of traditional agricul- tural exports. In the early 1970s, a balance of payments crisis resulted in a major devaluation of the peso. By then, the high economic cost of the import-substitution industri- alization strategy and the detrimental effects of the strategy on export potential were becoming well recognized. Nonetheless, the policy response was to provide industrial incentives directly to selected firms, including exporting enterprises, through tax holidays and the like (Bautista and Tecson 2003). No attempt was made to modify the highly protective tariff system. In fact, tariff protection was raised on many import-competing products, such as primary and processed food and agricultural products, chemical products, metal manufactures, electrical appliances, machinery, and transport equipment. During this period, the Philippines had the highest average tariff rate in Southeast Asia (Intal and Power 1991). In addition, the share of imported products (based on the seven-digit Philippine Standard Industrial Classification) subject to quantitative restrictions rose from 26 percent in 1970 to 52 percent in 1980 (Bautista and Tecson 2003). In the early 1980s, the government adopted various structural adjustment and stabilization measures to correct fundamental distortions in economic incentives and imbalances in the external and public sector accounts. These measures included the liberalization of the foreign exchange market, as well as trade policy reforms to remove quantitative trade restrictions and reduce the level and disper- sion of tariffs. The tariff reform program--the first of the unilateral trade liberal- ization programs--was instituted in 1981 to respond to a condition for a World Bank structural adjustment loan package. Under the program, the dispersion of tariff rates was to be reduced by lowering the peak tariff rates of 100 and 70 per- cent to 50 percent in two stages, while the low tariff rates were to be raised to at least 10 percent by 1985. Overall, the average tariff dropped substantially, from 43 percent in 1980 to 28 percent in 1985. To complement the tariff reform pro- gram, import licensing was also to be relaxed gradually. The plan was to remove selected nontariff barriers over three years. From the original list of 1,300 import items banned or requiring prior approval by the Central Bank of the Philippines and other government agencies, 264 items were to be removed in 1981, another 610 in early 1982, and the remainder by the end of 1983 (Bautista and Tecson 2003). However, the balance of payments crisis of 1983, following Benigno Aquino's assassination, stalled these initiatives. The government again adopted comprehensive 232 Distortions to Agricultural Incentives in Asia import and foreign exchange controls, rendering the tariff reductions ineffective. Commercial banks were required to turn over their foreign exchange receipts to the central bank so that priority imports and other payments could be facilitated. A 5 percent general import tax was imposed in November 1983 to generate gov- ernment revenue and discourage imports; the rate was raised to 8 percent in April 1984 and then to 10 percent two months later. Additional export duties ranging from 2 to 5 percent were levied on traditional export products from November 1983 to December 1984, and an economic stabilization tax of 30 percent was levied on all exports briefly in 1984. To curtail imports and capital outflows, the peso was devalued on successive occasions, and, in late 1984, the exchange rate was allowed to float. Tax reforms in 1983­85 gradually unified the sales tax on imports and import substitutes, remov- ing one source of import protection. The markup rate (which increases the tax base for imports) on essential and semiessential goods was reduced to a uniform 25 percent in 1985 and removed altogether in 1986. Under the new government of Corazon Aquino, the trade liberalization pro- gram was revived in 1986. Export taxes on all commodities except logs were abol- ished. The process of lifting import licensing was accelerated: 951 import items were liberalized in 1986, 170 in 1987, and another 209 in 1988. Of the remaining 673 restricted import items, those on list A were scheduled for immediate liberal- ization, and 94 had been liberalized by the end of 1989; those on list B were sched- uled for review. However, those on list C (114 items) continued to be restricted for national security or health reasons. The second tariff reform program was launched in July 1991 through an exec- utive order. This was intended to reduce the range of tariff rates to 30 percent over a five-year period. Although about 10 percent of all commodity lines were still subject to tariffs outside the target range, the average tariff rate had declined from 33 percent in 1990 to 27 percent by the end of 1995. As part of the second tariff reform program, an executive order was also issued to convert the quantitative restrictions on 153 agricultural products into tariff equivalents and realign the tariffs on 48 commodities. However, the order was soon reversed based on the Magna Carta of Small Farmers, which was passed in 1991 and, among other provisions, allowed the government blanket authority to restrict agricultural imports that competed with domestic production. The price and trade protection supplied for most major import-competing agricultural products were largely untouched by the series of unilateral trade lib- eralization measures introduced beginning in the late 1980s, however. Despite World Trade Organization (WTO) and regional trade agreements, nontariff trade barriers continue to distort the prices of some of the most important commodities. The Philippines 233 Agriculture-specific policies up to the mid-1980s Although they may be generally levied on all agricultural products and inputs, import tariffs are more commonly applied only on agricultural inputs and agri- cultural products that are not locally produced in any significant quantity. Such products include milk, wheat, and soybeans. Tariff protection for nontraded com- modities and for exportable products that are competitive in world markets is redundant because of prohibitive transport and other marketing costs. Over the period under study, a wide variety of policy instruments that influ- ence price incentives were applied to major agricultural commodities. These included government monopoly control over international trade and domestic marketing operations, import bans, quantitative trade restrictions, import licens- ing, export taxes, and export bans.3 Furthermore, despite serious efforts at unilat- eral trade liberalization, some of the policy instruments applicable to the more important import-competing agricultural products were kept largely intact. Indeed, some of the interventions had a long history going back to the U.S. Com- monwealth period that began in 1935. During the martial law era under Marcos, which began in late 1972, the government had considerably more leeway in inter- vening in setting prices and in marketing in agriculture, and it exercised this power. In the following discussion, we therefore distinguish between the evolution of agricultural price intervention policies before and after 1986, when the Cora- zon Aquino administration took over the reins of government. Rice and maize When bad weather caused a drastic shortfall in staple food grains in 1936, the National Rice and Corn Administration was established to ensure low, stable prices for consumers and adequate price incentives for farmers. To achieve these conflicting objectives, the agency was granted monopoly control over the imports and exports of rice and maize, as well as budgetary support and a credit line to undertake domestic market operations to defend price floors and retail ceilings and narrow the geographical and seasonal dispersion of prices. Because of high world commodity prices in the early 1970s, the government monopoly control over food commodities through this agency, which had been renamed the National Food Authority, was expanded beyond rice and maize to allow the tariff-free importation of wheat, soybeans, soybean meal, ruminant live- stock, and beef. Sugar The sugar industry has historically been assisted more than any other industry. This is because of the preferential access assigned to Philippine sugar in the U.S. market beginning in 1902 and the authority entrusted to the Philippine 234 Distortions to Agricultural Incentives in Asia government to administer its sugar export quota to the United States through the Jones-Costigan Act passed by the U.S. Congress in 1934. Initially, the domestic quota system was established for the orderly distribution of the U.S. quota among sugar producers. In the 1960s, the quota system was also charged with reducing the burden on domestic consumers of the higher export prices resulting from the 1962 devaluation and the greater U.S. quota allocation arising because of the Cuban missile crisis (given Cuba's role as a major pro- ducer). Under this system, which is still in force, producers are paid a composite price derived as an average of the export price, a lower domestic wholesale price, and a reserve price weighted by the quantity allocations targeted for U.S. exports, the domestic market, and as a reserve. When the Laurel-Langley Agreement--a trade agreement between the United States and the Philippines--expired in 1974, sugar trading was effectively nation- alized, first under Philippine Exchange Inc. and subsequently under the National Sugar Trading Corporation. The latter was the sole Philippine wholesale buyer and seller of sugar in both the domestic and international markets until the end of the Marcos administration in early 1986. It also established new refineries, oper- ated sugar market outlets, and acquired leading enterprises involved in the trans- port, storage, and handling of sugar for export. Export commodities Except in the aftermath of the 1970 devaluation and the sharp increases in world commodity prices in the mid-1970s, there have been few attempts to intervene in the production and trade of exportable agricultural products. Initially, as part of stabilization measures, export taxes from 4 to 6 percent were imposed on major agricultural exports and other primary exports; to generate revenue, these taxes remained in place until the mid-1980s. The higher rate of 6 percent was imposed on traditional exports of coconuts (copra) and centrifugal sugar to promote more processing of agricultural exports. The lower rate of 4 percent was applied to coconut oil, desiccated coconuts, coconut cake and meal, molasses, abaca, bananas, and pineapple products.4 In 1974, because of the world commodity boom, additional premium duties were briefly imposed on exports. The duties ranged from 20 to 30 percent of the difference between the ruling export price and a February 1974 base price. Wind- fall gains from the devaluation and the commodity boom were thus partially siphoned off from the producers of these agricultural and primary exportable commodities. The Coconut Consumers Stabilization Fund implemented an additional levy on the coconut industry in 1973 (commonly known as the coco levy). The levy was aimed partly at protecting domestic consumers from a sharp rise in the price The Philippines 235 for coconut oil on the world market and partly at raising funds for the develop- ment of the coconut industry.5 There was also a belief that taxing or restricting coconut exports might be beneficial because the Philippines was thought to pos- sess some monopoly power on the world market. (While the Philippines accounted for a high share of total world coconut and coconut oil exports, any monopoly power was curtailed by the fact that coconut oil comprises only about 7 percent of the world lauric oil market.) Some of the revenue from this levy was used to buy up 80 percent of the coconut oil milling industry, which was then reorganized under a newly created and privately owned company, United Coconut Oil Mills. This company eventually acted as a monopsonist buyer of coconut in the farm sector. When the world prices of coconut oil fell in 1982, the levy was lifted, only to be replaced by a ban on coconut exports to protect coconut oil mills. Agriculture-specific policies after 1986 World commodity prices began to fall in the late 1970s. However, the policies and institutions established to cope with high world prices persisted because they proved to be a convenient means of raising revenues. The related measures heavily taxed farmers so as to support private interests and cover bureaucratic inefficiencies (David 1983). In 1986, under the new Corazon Aquino govern- ment, several of the direct government price and market regulations were finally dismantled. The reforms under Corazon Aquino Export taxes, including the coconut export ban, were abolished, and the National Food Authority's monopolistic controls over international trade in wheat, soybeans, soybean meal, and meat were removed, limiting the authority's functions to rice and maize (also the case prior to martial law). The authority's domestic marketing operations were also effectively reduced because financial support had to be provided mostly through annual budgetary allocations rather than profits from imports. The National Sugar Trading Corporation was eventually replaced by the Sugar Regulatory Administration, which functioned primarily as a market regulator. It mainly allocated U.S. sugar quotas, determined the allowable quantities of imports, the importers, the amounts to be sold in the domestic market, and the amounts to be kept in reserve. It also performed developmental functions, such as research and extension, but it has ceased to engage in direct market operations. Quantitative trade restrictions on fertilizers were removed, and tariffs on major agricultural inputs were lowered substantially. However, the monopsonist 236 Distortions to Agricultural Incentives in Asia control exercised by United Coconut Oil Mills over the coconut market has con- tinued; the government's attempts to wrest ownership of this entity have been bogged down in court proceedings. Despite efforts at trade liberalization in the late 1980s, most major importable agricultural commodities that are also produced in significant volumes in the Philippines are still subject to quantitative import restrictions, particularly those commodities protected by laws passed by Congress. Efforts to remove the quanti- tative restrictions were preempted by the passage of the Magna Carta of Small Farmers in 1991, which provided blanket authority to restrict agricultural imports that competed with domestic production. In addition, the Seed Law was passed to regulate imports of seeds and other planting materials. The Uruguay Round Agreement on Agriculture The country's ratification of the Uruguay Round Agreement on Agriculture in the mid-1990s promised to set a decisive path toward trade liberalization in the sec- tor. It aimed to replace all quantitative restrictions by tariffs, impose a ceiling on tariff rates, and reduce tariff protection. Unfortunately, the specific agreements and the manner of implementation did not live up to the promise (David 1994, 2003). First, one of the most heavily regulated commodities, rice, was exempted from tariffication until 2004, which is similar to the situation in Japan and the Republic of Korea.6 Second, the quantitative trade restrictions lifted by executive order in April 1996 were replaced by tariff rate quotas that initially raised out-of-quota tariffs to the maximum (bound) tariffs allowed under the WTO commitments, while the in-quota tariffs were set mostly at the levels existing in 1995. The initial out-of- quota tariffs of 100 percent were typically higher than the nominal protection rates implied by the quantitative restrictions in 1990­94. They were also higher than the book tariff rates under the executive order that set out the unilateral tar- iff reductions on a wide range of agricultural and industrial goods. Despite the reductions in the out-of-quota tariffs scheduled for no later than 2004, the tariff rates are still equal to or higher than the tariff rates in 1995, and they are higher than the government's target average tariff of 5 percent. Furthermore, whereas the quantitative restrictions on primary and lightly processed products were supposed to be lifted, the tariffs were raised on a number of imported agricultural products that are considered substitutes for other com- modities (for example, feed wheat and barley as substitutes for maize). Tariffs were also increased to this same level on more heavily processed products using these commodities as main raw materials (for example, preserved and canned meat products, milled rice and maize, and roasted coffee). The Philippines 237 Third, the manner in which the minimum access volume provision or the tariff rate quota system of the Uruguay Round Agreement on Agriculture was adminis- tered for major import-competing commodities simply perpetuated quantitative trade restrictions. The quantities that may be imported at the lower in-quota tariffs were sometimes changed to prevent domestic prices from rising sharply whenever production shortfalls occurred. Thus, in effect, the tariffs were no different than quantitative restrictions. For certain commodities, the right to import was assigned mostly to domestic producers of the same products (for example, pork and poul- try), and these producers have often chosen not to utilize their import allocation so as to protect their domestic production. Moreover, because most of the minimum access volumes were lower than the demand for imports at the in-quota tariff rate and because the right to import the minimum access volumes was not auctioned, large quota rents accrued to those entities given access to the minimum access vol- ume allocation, at least during the early period of implementation. Finally, the agreement's lack of provision for the market operations of paras- tatals allowed the Sugar Regulatory Administration to continue exercising its reg- ulatory authority over import levels and the market destinations of domestic sugar production. In fact, the domestic market operations of the National Food Authority in support of producer prices were expanded in the late 1990s to include sugar.7 Regional trade agreements In 1992, the six member countries of the Association of Southeast Asian Nations at the time (Brunei Darussalam, Indonesia, Malaysia, the Philippines, Singapore, and Thailand) agreed to form a free trade area. The aim was to reduce tariffs to between 0 and 5 percent and to abolish quantitative trade restrictions and other nontariff trade barriers by 2010. Under the related Common Effective Preferential Tariff Scheme, unprocessed agricultural products were commonly included in a sensitive products list that allows trade liberalization to begin later (between 2001 and 2003), though the 0 to 5 percent tariff targets are still to be reached by 2010. A recent study has indicated, however, that the effective protection rates in agricul- ture have been largely unaffected by the scheme thus far (Pimentel 2006). Unlike other free trade areas, which usually delay or totally exclude trade liber- alization in agriculture, the agreement signed between China and the free trade area of the Association of Southeast Asian Nations in 2002 specifically covered a significant portion of all agricultural products through the Early Harvest Program. The agreement called for the elimination of tariffs on live animals, meat and edible meat offals, fish, dairy products, other animal products, live trees, vegetables, fruits and nuts, and a few commodities in other categories of the Harmonized Commodity Description and Coding System. The changes were to 238 Distortions to Agricultural Incentives in Asia be implemented starting in January 2004 and were to be in full operation no later than January 2006 in the free trade area, although exceptions were allowed in some cases depending upon negotiations with China. The Philippines signed an Early Harvest Program Agreement with China in early 2006. The agreement covered proportionately fewer commodities than the agreements reached by other countries of the Association of Southeast Asian Nations, suggesting that there is less interest in the Philippines in engaging with China in more open agricultural trade (Pasadilla 2006). In general, consumers will benefit from lower prices on covered commodities, while producers of exportable bananas, pineapples, mangoes, other tropical fruits, and coconut and coconut oil are expected to gain from greater access to the vast Chinese market. However, pro- ducers of vegetables, leguminous crops, and pig meat, together with producers of fruits for local markets, will be hurt, at least initially, by the entry of relatively cheap frozen meat and other meat products, potatoes, carrots, onions, garlic, peanuts, pears, apples, oranges, and other fruits and vegetables from China. Estimates of Nominal and Relative Rates of Assistance Measuring the distortions in incentives caused by price and trade policies has had a long history in the Philippines. The first effort was undertaken by Power (1971) in a study on the situation in 1965. This was followed by studies conducted by Norma Tan (1979), on 1974; Elizabeth Tan (1994), on the 1980s; Manasan and Pineda (1999), on the 1990s; and Aldaba (2005a, 2005b), on recent years. These studies estimated the effective protection rates in all industries, both agricultural and nonagricultural. However, the main interest and analysis were concentrated on the manufacturing sector. The first studies on agricultural protection were conducted by David (1983), Intal and Power (1991), and, more recently, David (2003). Unlike the industrial protection studies that quantified the effects of tariffs and indirect sales taxes, these agricultural protection studies were based on domestic and border price comparisons. This enabled the effects of nontariff trade barriers to be measured and redundancy in the tariffs, if any, to be taken into account. The indirect impacts of industrial protection and other economy-wide factors on agricultural incentives (through effects on the exchange rate) were also analyzed. Methodology In this study, we estimate nominal rates of assistance (NRAs) in industries. The main focus of our study's methodology is government-imposed distortions that create a gap between domestic prices and the prices that would emerge under free The Philippines 239 Figure 6.1. Value Shares of the Primary Production of Covered and Noncovered Commodities, the Philippines, 1966­2004 100 90 80 70 prices 60 50 distorted 40 at 30 %, 20 10 0 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 year residual bananas beef chicken coconuts maize pig meat sugar rice Source: Author calculations based on producer price and production data in FAOSTAT Database 2008. market conditions. (For a detailed description of the methodology, see Anderson et al. 2008 and appendix A.) Since it is not possible to understand the characteris- tics of agricultural development through a sectoral view alone, our project's methodology not only estimates the effects of direct agricultural policy measures, but, for comparative evaluation, also generates estimates of the distortions in nonagricultural sectors. More specifically, our study computes NRAs for producers of the main farm products (figure 6.1). It also generates NRAs for nonagricultural tradables for comparison with the NRAs for agricultural tradables via the calculation of a rela- tive rate of assistance (RRA). This approach provides a consistent time series measure of distortions over more than four decades using the values of produc- tion as weights to compute sectoral and subsectoral averages. This contrasts with the previous studies, which, except for Manasan and Pineda (1999), computed sectoral averages using trade volumes as weights. We assume that the Philippine economy is small and open and, thus, that the country's level of trade does not affect world prices.8 Border prices are estimated based on world price series reported by the World Bank. For importables, these data are adjusted to values that include cost, insurance, and freight by assuming the cost of transport and insurance at a constant 20 percent of free on board world 240 Distortions to Agricultural Incentives in Asia prices. We have not chosen the country's officially recorded unit values for imports or exports because foreign exchange controls, export taxes, and other taxes mean that the export unit data are significantly undervalued, particularly up to the mid-1980s. For rice, maize, and sugar, the import unit data may be either overvalued (in the case of imports by the National Food Authority) or underval- ued (because private importers seek to lower their tariff payments by underin- voicing). There were also no imports of some importable products during various periods. Rice, maize, sugar, pig meat, beef, and poultry are consistently classified as importables even though there were no imports of these products during some years. In the case of sugar, the first imports occurred only in the 1990s, but, even in the 1960s and 1970s, exports of sugar were confined to the high-priced U.S. mar- ket, where the Philippines has had preferential access, and were not competitive at the free-market world price. Sugar is thus also treated as an importable. In the absence of detailed time series data on marketing costs, we define the domestic price as the wholesale price that is the closest to the same point of the border price in the marketing chain. For most of the agricultural commodities for which we compare the domestic and border prices, we rely on the data for the commodity that is internationally traded and is lightly processed rather than the data on the primary product sold at the farmgate. Thus, we take milled rice versus paddy; raw or refined sugar versus sugarcane; frozen pig meat, beef, or poultry versus hog, cattle, or chicken birds. In the case of sugar, the rates of the protection received by farmers and millers are the same because the revenues derived from the sale of raw and refined sugar in both the domestic and U.S. markets are shared proportionately between the two at the ratio of 70 to 30.9 The ratio of the farm price of paddy to the retail price of rice did not change significantly, suggesting that farmers and rice millers, together with traders, share proportionately in the protection accorded the rice industry. For rice, maize, pig meat, beef, and poultry, we assume that the NRAs of the processed prod- ucts and the farm products are equivalent. Indeed, the import tariffs on the farm products are generally the same as the tariffs on the lightly processed varieties. Aside from the major agricultural commodities specified above, we have esti- mated the NRAs for other, noncovered crops within the sector. Since price com- parisons are more difficult to perform in the case of noncovered exportables, including pineapples, mangoes, abaca, and tobacco, we have assumed that the relevant average NRAs are zero or equal to the export tax whenever this applies. For products that are nontraded because of prohibitive marketing costs, such as roots and tuber crops, we have assigned zero NRA values.10 For the many import- competing vegetables, fruits, and other minor crops, we assume that the NRAs for the group are the same as the average NRAs for covered importable products.11 The Philippines 241 We also assume that the weights are one-third each for exportables, importables, and nontradables in the noncovered part of farm production (which, in aggre- gate, amounts to around one-fifth of the agricultural sector's value of production at undistorted prices). For importable nonagricultural products, the NRAs are generally based on book tariff rates, apart from the case of lightly processed food manufacturing industries in which the NRAs are based on price comparisons, including rice and maize milling, sugar milling and refining, coconut oil production and refining, and so on.12 For a number of nonagricultural primary products in fishing, forestry, and mining, we use the export taxes that applied from 1970 to 1985. The definitions we use for a product's tradability and for industry weights are the same for the agricultural and nonagricultural sectors. NRAs in agriculture Our estimates of the NRAs for agricultural commodities from 1962 to 2004 are summarized in table 6.5. Although the estimated NRAs are highly variable over time (see the annual estimates in David, Intal, and Balisacan 2007), several general patterns emerge from the five-year averages shown in the table. First, import-competing products have enjoyed much more assistance than exportables. Coconut production has been penalized by negative NRAs over the entire period of our study, averaging around 20 to 25 percent from the 1970s to the mid-1980s. Besides the multiple exchange rate policy prior to 1970 (not measured here), this was caused by the imposition of several measures, including the export tax, the coconut levy, and the coconut export ban that were aimed at siphoning off windfall gains from the 1970 devaluation; the subsequent world commodity boom of 1973­74 that was incompletely transmitted to the domestic market; and the lower raw material costs for the coconut oil milling industry. Despite the abolition of these policy instruments in 1986, however, coconut farm- ers continued to be implicitly taxed, albeit at a lower rate of around 15 percent of border prices. Evidently, the government's failure to dismantle the ownership of 70 to 80 percent of the coconut oil milling industry by United Coconut Oil Mills has allowed this entity to maintain its monopsonist power over the domestic price for coconuts even now. Second, among import-competing commodities, the level of the NRAs has dif- fered significantly, and the differences have widened between two groups of import-competing products: the NRAs for the most important commodities-- rice, maize, and sugar--have increased, while those for the many more minor, but higher-valued commodities have declined. Each of these commodities is consid- ered in detail below. 242 Table 6.5. NRAs for Covered Agricultural Products, the Philippines, 1962­2004 (percent) Indicator, product 1962­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Exportablesa 9.0 7.6 14.3 14.6 22.6 16.5 11.4 5.4 8.7 Coconuts 24.4 20.2 25.3 16.7 27.1 20.6 15.3 7.8 14.1 Bananas 0.0 0.0 4.0 4.0 4.0 0.8 0.0 0.0 0.0 Import-competing productsa 1.1 17.4 5.6 5.8 2.2 30.2 25.1 48.1 31.4 Rice 3.4 1.4 9.7 17.9 16.3 14.5 20.9 52.7 50.7 Maize 0.5 38.4 14.0 24.3 20.1 59.8 62.6 78.5 54.5 Sugar 8.2 120.7 11.7 1.7 59.5 123.2 49.3 97.2 79.3 Beef 15.0 15.0 12.0 10.0 5.0 17.0 28.0 28.0 10.0 Pigs 30.0 13.6 3.2 5.5 35.8 51.0 25.1 20.6 8.3 Chicken 8.9 67.1 28.9 28.1 38.4 42.9 56.5 42.2 52.1 All covered productsa 1.9 15.3 6.5 8.1 5.1 16.1 17.5 37.9 24.9 Dispersion, covered productsb 17.1 29.6 25.2 22.3 28.6 29.9 27.5 27.9 30.4 Coverage, at undistorted prices 78 79 79 74 73 73 77 77 80 Source: David, Intal, and Balisacan 2007. a. Weighted averages; the weights are based on the unassisted value of production. b. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of covered products. The Philippines 243 Third, the increasingly higher NRAs observed since the 1980s may be partly caused by the government's efforts to reduce the burden of the adjustment of the agricultural sector to a long-term decline in world commodity prices. This declin- ing trend is visible in real domestic prices relative to real world prices for several commodities noted in David, Intal, and Balisacan (2007). Fourth, the dispersion of NRAs among agricultural products within the farm sector (measured according to the standard deviation of these NRAs and reported in table 6.5) has not diminished. The trade bias index has not diminished either (table 6.6), indicating that the NRAs for importable farm products have persist- ently remained above the NRAs for exportables. Both of these indicators imply that the efficiency of resource use within the farm sector has been substantially compromised by agricultural policies. Fifth, the NRAs have fluctuated from year to year, mostly in response to world price changes and sometimes in response to exchange rate adjustments (see fig- ure 6.2). For example, the NRAs for import-competing agricultural products were below the trend in 1973­74 and 1980, when international prices were high, but above the trend in the mid-1980s, when international prices were low. This sug- gests that domestic price stabilization has been an important objective of agricul- tural price and trade policy. Table 6.7 shows that the estimated coefficients of the variation in domestic prices tended to be considerably lower than the correspon- ding coefficients in world prices, particularly among major import-competing commodities. Rice The trends in the NRAs for rice reflect the inability of the National Food Author- ity to attain simultaneously its inherently conflicting objectives of providing low prices to consumers and remunerative incentives to farmers. Prior to the late 1980s, the domestic price of rice was, on average, about equal to the long-term level of the border price. The negative protection from the 1970s to the early 1980s was caused by the unusually high world prices during this period. This did not discourage farmers, however, because the Green Revolution in rice and the land reform in rice farming were transforming tenant farmers to owner- operators.13 Because of the drop in the world price of rice, a sharp fall in irrigation invest- ments, and stagnation in the yield potential of newer modern varieties, growth in the demand for rice increased more rapidly than rice production beginning in the late 1980s. The NRAs for rice became positive; they had risen to about 50 per- cent by the early 2000s. This occurred despite the most significant rice imports ever, which reflected the country's rising comparative disadvantage in rice pro- duction. 244 Table 6.6. NRAs in Agriculture Relative to Nonagricultural Industries, the Philippines, 1962­2004 (percent) Indicator 1962­64 1965­69 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Covered productsa 1.9 15.3 6.5 8.1 5.1 16.1 17.5 37.9 24.9 Noncovered products 0.4 5.8 2.3 2.4 0.3 10.0 8.4 16.0 12.3 All agricultural productsa 1.6 13.3 5.6 6.6 3.6 14.4 15.4 33.0 26.0 Non-product-specific assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total agricultural NRAb 1.6 13.3 5.6 6.6 3.6 14.4 15.4 33.0 26.0 Trade bias indexc 0.03 0.18 0.04 0.03 0.15 0.31 0.26 0.34 0.31 NRA, all agricultural tradables 1.7 14.3 6.0 7.2 4.0 15.8 16.7 35.7 27.9 NRA, all nonagricultural tradables 19.0 20.3 16.3 16.3 12.9 11.0 9.9 8.6 7.3 RRAd 17.4 5.0 19.8 20.3 14.9 4.3 6.1 24.9 19.1 Source: David, Intal, and Balisacan 2007. a. Including product-specific input subsidies. b. Including product-specific input subsidies and non-product-specific assistance. The total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. c. The trade bias index is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. d. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. The Philippines 245 Figure 6.2. NRAs for Exportable, Import-Competing, and All Agricultural Products, the Philippines, 1962­2004 60 50 40 30 20 % 10 NRA, 0 10 20 30 40 1962196419661968197019721974197619781980198219841986198819901992199419961998200020022004 year import-competing products total exportables Source: David, Intal, and Balisacan 2007. Maize Maize is a food staple for 10 to 15 percent of the population and a major feed ingredient for livestock. In contrast to rice, however, domestic production in maize has been consistently protected. The NRAs for maize grew steadily from about 25 percent in the late 1970s and early 1980s to nearly 80 percent in the late 1990s. Unlike rice, there is less political pressure to lower maize prices for poor consumers because maize is mostly a subsistence crop among upland farmers in the southern part of the country. Sugar If sugar is categorized as an importable and the world price of sugar--not the unit value of exports to the premium U.S. market--is used as the border price in the cal- culation of NRAs, then the domestic sugar industry is clearly the most highly pro- tected industry throughout the period we study. U.S. consumers paid for a large share of the income transfers to the sector in the 1960s and early 1970s when nearly all domestic production was exported. However, the burden shifted to consumers and food processors in the Philippines when the Laurel-Langley Agreement expired in 1974. At that time, the U.S. sugar quota dropped sharply. Exports, which continue to be confined to the protected U.S. market, now account for only around 10 percent of domestic production. Yet, because of import restrictions, the average NRA in the sugar industry has increased, averaging around 90 percent during the past decade.14 246 Distortions to Agricultural Incentives in Asia Table 6.7. Coefficients of Variation in Real International Prices and Philippine Wholesale Prices, Major Agricultural Commodities, 1960­2004 Product 1960­2004 1960­80 1980­2004 Rice World 47 32 34 Domestic 19 12 11 Maize World 38 17 27 Domestic 23 17 14 Coconuts World 49 27 40 Domestic 45 31 36 Coconut oil World 50 27 43 Domestic 43 32 33 Sugar World 80 71 60 Domestic 19 14 18 Beef World 33 21 23 Domestic 28 35 17 Pig meat EUV Sing 46 44 23 Domestic 13 12 14 Poultry EUV Sing 37 33 14 Domestic 29 17 15 Source: Author estimates based on data in David, Intal, and Balisacan 2007. Poultry and other livestock products Poultry producers benefited from a high level of protection. A slightly rising trend in the NRAs has been discernable, from about 40 percent prior to 1985, when the high tariff protection (70 percent) was redundant to a significant extent, to about 50 percent thereafter. The tariff protection for the pig meat industry has been significant, although it has generally been lower than the protection for poultry. Beginning in 1995, the govern- ment adopted the same level of high in- and out-of-quota tariffs for both poultry and pig meat under the Uruguay Round Agreement on Agriculture. However, the low tar- iff was largely redundant up to the 1980s, as were the relatively high tariffs after 1995. The tariff protection for beef has been historically less than that for pig meat. The NRAs for beef averaged around 10 percent until the late 1980s. Beef was not The Philippines 247 included among the sensitive products on which tariffs were raised following the WTO agreement. Nonetheless, an upward trend in the estimated NRAs for beef may also be observed for the 1990s. It appears that the government's attempt to promote activities aimed at fattening cattle, which relied on allowing duty-free imports of young cattle from Australia, was accompanied by more restrictive non- tariff trade barriers on beef to increase incentives. The expansion of the cattle fat- tening business was short-lived, however; tariffs on beef were reduced in the late 1990s, and import restrictions became untenable.15 That beef benefited from less protection than poultry may be explained by the prevalence of large-scale integra- tion farming operations that produced poultry through contract farming, in con- trast to the beef industry, which is dominated by backyard producers. However, the pig meat industry is also dominated by backyard producers, and this subsector benefited from relatively high protection rates. Agricultural inputs To infer the effect of price interventions on value added, David, Intal, and Balisacan (2007) report on trends in the NRAs for the major intermediate inputs commonly used in agricultural production. Until the mid-1980s, the govern- ment's policies on industrial promotion raised the domestic prices for manufac- tured inputs in agriculture significantly. The consumer tax equivalent of the import protection provided for agricultural inputs such as fertilizers, agricultural chemicals, farm machinery, and even water pumps was generally higher than the NRAs for agricultural outputs, aside from the case of sugar in some years. With the exception of subsidies for gravity irrigation in rice, there was no significant offsetting input assistance in agriculture. Indeed, despite price controls, tax free imports, and direct subsidies for fertilizer companies before the mid-1980s, the consumer tax equivalent on farm inputs was negative only during 1970­74 (because of the fourfold jump in world oil prices). These data are not incorpo- rated in the aggregate NRAs for crop agriculture; but, if they had been incorpo- rated in the manner described in Anderson et al. (2008) and appendix A, they would have lowered the estimated NRAs for crops by a few percentage points, especially in the earlier decades of the period we study. The estimated NRAs for livestock would also be lower if the impact on the prices of feedmix deriving from the import restrictions on maize had been taken into account. NRAs for nonagriculture and RRAs for agriculture Agricultural incentives are affected indirectly by rates of assistance to nonagricul- ture. Mobile resources move across sectors or industries according to the relative incentives. Figure 6.3 and table 6.7 show the trends in the average NRAs of the agricultural and nonagricultural sectors, as well as the RRAs for agriculture. 248 Distortions to Agricultural Incentives in Asia Figure 6.3. NRAs for Agricultural and Nonagricultural Tradables and the RRA, the Philippines, 1962­2004 50 40 30 20 10 cent 0 per 10 20 30 40 50 1962196419661968197019721974197619781980198219841986198819901992199419961998200020022004 year NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: David, Intal, and Balisacan 2007. The contrast in the trends between the two sectors is striking. Whereas the average NRAs for agriculture were lower than the average NRAs for nonagriculture prior to the mid-1980s, they have risen to more than 30 percent in the past 10 years or so. Meanwhile, because of unilateral trade liberalization measures, the NRAs in nonagriculture declined steadily from nearly 20 percent in the 1960s and 1970s to only 7 percent by the early 2000s. As a consequence, the assistance to agriculture is now much greater than the assistance to nonagriculture; thus, RRAs have risen from an average of around 15 percent prior to the mid-1980s to an average of more than 20 percent in recent years. This trend indicates that the efficiency of resource use in farming rel- ative to the production of nonagricultural tradables first increased as RRAs became less negative and then decreased as RRAs became more positive. This indicates that there were too few resources in the agricultural sector up to the mid-1980s, but that, since then, there have been too many resources, on average (especially in import-competing agriculture). Explaining the Patterns in the Distortions in Agricultural Incentives In general, countries have shifted from taxing to subsidizing agriculture in the course of economic development primarily because of political economy factors (Anderson and Hayami 1986; Lindert 1991; Anderson 1995). In any country that The Philippines 249 would be self-sufficient in food in a world of free agricultural trade, this shift is expected to occur when the country's per capita income reaches 2.6 times the global average (Tyers and Anderson 1992). In a country that would be only 65 percent self-sufficient in food under free trade, the shift would occur when the per capita income reaches the global average (US$4,300 in 1992). In the case of the Philippines, the shift from taxing to assisting agriculture directly through price interventions occurred at a level of economic development that was lower than predicted by earlier studies: per capita income in the Philippines was only about US$1,200 even in the late 1990s. Why did the switch toward higher agricultural protection occur early in the Philippines? The explanation lies in unique historical events, political economy factors, the local political system, and the strong nationalist sentiment for food self-sufficiency, especially in rice. The country's highly skewed distribution of landownership and dualistic agrarian structure that arose from colonial land poli- cies and agroecological conditions meant that there were large landowners and plantation operators who were able to lobby effectively for their own interests. The farm producers in these groups provided the major political leaders in Con- gress and in the executive branch of government at all levels. Landed oligarchs also represented the business elite, which successfully pushed for an industrial protec- tion policy that biased incentives against the agricultural sector (Hara 1994). As the international pressure for trade liberalization has mounted during the past two decades, it has become easier to resist the opening of domestic markets for food staples by playing up the national sentiment for food self-sufficiency. Politi- cal pressure to raise agricultural protection has been strengthened by expanded lobbying efforts undertaken by farmer organizations, large landowners, and agribusiness firms such as livestock and poultry producers, millers, seed compa- nies, and input suppliers. There has been little resistance to the high prices for white maize as food because white maize is primarily a subsistence crop. Livestock producers (including poultry producers) and feed millers who use mostly yellow maize have chosen to lobby for greater protection for livestock output to offset the high maize prices rather than lobby for a more rational policy in maize and livestock. The objections to the highly restrictive policy on maize imports have been addressed by providing import alloca- tions at lower tariffs for the large, more well organized, and more vocal feed, poultry, and pig meat industries. The larger feed mills and livestock producers also own flour mills and are thus able to substitute low-grade wheat, which is subject to only a 10 percent tariff, for the artificially high-priced maize. This policy structure provides large-scale feed and livestock producers with a cost advantage over smaller produc- ers who must rely on the domestic market for maize supplies. Aside from the large size of farms and mills, the sugar subsector has historically had strong political power because of its close relationship and common interest 250 Distortions to Agricultural Incentives in Asia with the government in lobbying to protect the country's preferential market access to the U.S. sugar market. In contrast, the share of sugar in direct household expenditure is small, and consumers have therefore generally tolerated or been unaware of the high sugar prices. Resistance from the food processing subsector against high sugar prices was mitigated by granting larger, more vocal food processors the privilege to import sugar free of tariffs. Concluding Remarks Price intervention policies became more favorable in the agricultural sector beginning in the mid-1980s. The protection for major import-competing com- modities was increased, and unilateral trade liberalization measures lowered the implicit tariffs on inputs and the protection of nonagricultural sectors. Thus, improvements in agricultural incentives occurred at the cost of the inefficiencies in resource allocation arising from widening distortions in prices within agriculture and between agriculture and nonagriculture. Artificially raising the profitability of major import-competing commodities directly increased the cost of land for the production of other crops. It also indi- rectly reduced the competitive advantage of exportable agricultural products in world markets. The high-price maize policy also lowered the international com- petitiveness of the pig meat industry, in which the Philippines may well have a comparative advantage. The significant protection for sugar hurts not only con- sumers, but also the food processing industry, which accounts for over 20 percent of manufacturing employment and value added. The excessive protection for major staple food commodities reduces the welfare of rural landless and urban poor households and puts pressure on wages, rendering labor-intensive manufac- turing industries less competitive relative to low-wage, low-cost food economies such as China and Vietnam. The economic waste caused by price intervention policies is magnified by the continued use of quantitative trade restrictions instead of tariffs. In particular, the government's monopoly on rice imports and domestic marketing operations through the National Food Authority has been extremely costly, and it has also failed to achieve the basic and conflicting objectives of lowering food prices for consumers, raising producer prices, and stabilizing both sets of prices. The use of quantitative restrictions promotes rent seeking, reduces government revenues, incurs significant bureaucratic costs, and aggravates price uncertainties. Unfortunately, recent policy changes in response to the WTO agreement on agriculture seem to have exacerbated rather than mitigated such problems; the nominal protection rates for major import-competing commodities have not only been raised, but the scope of the operations of the National Food Authority The Philippines 251 has also been inadvertently expanded. Rice market interventions and the use of quantitative restrictions have persisted because the economic costs and even some of the financial costs are not readily apparent to the general public. Meanwhile, the bureaucracy is being corrupted through commissions, bribes, and the other rents typically involved in government procurement and import licensing, mak- ing it even more difficult to effect trade liberalization. There are no indications that the government will move toward greater trade liberalization in agriculture in the near future. In the current negotiations under the Doha Round, efforts are being made to retain the relatively high level of tariff protection for major import-competing agricultural commodities. Furthermore, the government is not taking any steps to dismantle the institutions and other policy instruments that regulate the imports of rice, sugar, and other commodi- ties. Because of the sharp rise in world grain prices in recent years, the government will be politically compelled to continue its food self-sufficiency strategy rather than rely on open international markets to achieve food security. Notes 1. Because there was an acceleration in overseas labor migration, real wages did not decline. Migra- tion boosted foreign exchange earnings, and remittances began to account for at least a 10 percent share in gross national product. 2. Fishery products, led by tuna and shrimps, have become major agricultural exports. They now account for about 20 percent of total agricultural exports. 3. Even for relatively minor crops, specific laws were passed prohibiting imports. This affected, for example, onions, garlic, potatoes, and cabbages in 1955, coffee in 1960, and cigarettes and tobacco, except for blending purposes, in 1964. 4. Export taxes were also imposed on logs at 10 percent; copper ore at 6 percent; and other metal ores, shrimps, prawns, lumber, plywood, and veneer at 4 percent. 5. About 20 percent of the revenues from the tax briefly supported the direct subsidy on the domestic consumption of coconut oil products. The remainder was supposed to be used to finance development programs in the coconut industry such as replanting, vertical integration, and scholar- ships. Subsequent research showed that little benefit, if any, accrued to farmers from these expendi- tures (Clarete and Roumasset 1990). 6. The Philippines is still negotiating in the WTO's Doha Round to keep government monopoly control over rice imports. 7. The government also occasionally used the WTO-sanctioned safeguard mechanism when addi- tional tariffs were imposed on poultry. 8. The country is a significant trader only in coconut products, in which it has been the world's largest producer and exporter for most of the period we study. However, to a large extent, competing products such as palm oil and soybean oil may be substituted for coconut products. Coconut oil, the most important coconut product, constitutes only a small (7 percent) share of the world trade in vegetable oils. 9. According to Borrell et al. (1994), the quedan system, which allocates products to the various markets in fixed proportions, reduces the incentive to increase production and invest in yield- increasing technology. This is because higher production cuts into gross revenues. Also, because export allocation to the United States occurs at a fixed ratio, there is no incentive to improve milling quality for export so as to increase net returns. Meanwhile, the sugar-sharing arrangement--whereby 60 to 252 Distortions to Agricultural Incentives in Asia 70 percent goes to growers, and 30 to 40 percent goes to millers (depending on the recovery rate)--that was instituted by law to provide millers a share of the benefits from price protection also happens to reduce the incentive of growers and millers to raise productivity. Thus, growers receive only 60 to 70 percent of the benefits of productivity-enhancing investments, while millers receive only 30 to 40 percent of these benefits. 10. As described in Anderson et al (2008) and appendix A, a commodity is considered nontradable if the proportion of imports and exports in the total value of production is less than 5 percent. If a commodity or commodity group is both exported and imported in significant amounts (that is, more than 2.5 percent of the total value of output), we estimate the NRA as the average of the NRA of the commodity as an exportable and the NRA of the commodity as an importable, weighted by the respec- tive proportions of the export and import values to the total traded value. 11. Despite the supposed removal of quantitative trade restrictions on all agricultural commodi- ties except rice, nontariff barriers appear to be significant in the case of many commodities, such as vegetables, fruits, and meat. Price comparisons are difficult to perform for many of these products because of the lack of consistent world price series, difficulties in making adjustments for quality differences, and the complexity of measuring the effect the increased imports of a commodity that may not be grown in the country may have on the price of a highly substitutable product that is produced domestically. One clear indication of this effect, however, is the substantial smuggling of vegetables, fruits, and nuts from China. The exports of these commodities to the Philippines reported in Chinese trade statistics have been up to 10 times more substantial than the corresponding imports of these commodity groups reported in official Philippine trade statistics despite the already low tariffs on these smuggled products (mostly 3 percent, but, in some cases, up to 10 percent since the late 1990s). 12. To the tariffs in the computation of NRAs, we have added the differences between the indirect tax on domestically produced goods and the indirect tax on imported products imposed from 1960 to the early 1980s. While the tax rates on imports and the tax rate on exports were the same in most cases, the taxes were effectively greater in the case of imports because the tax base for imports was the tariff- inclusive price, augmented by a percentage markup. In 1974, the weighted average nominal tariff rate was 22 percent for manufacturing products, but the nominal rate of protection would nonetheless be significantly higher, 31 percent, if one takes into account the effect of the difference in the tax base (Tan 1979). 13. New seed fertilizer technologies and the accompanying expansion in irrigation increased the country's comparative advantage in rice production, briefly turning the Philippines from a net rice importer to rice self-sufficiency and, by the late 1970s, reducing the domestic price of rice in real terms. 14. Even the large margins conferred on sugar importers as a result of the high level of nominal protection have been received mostly by sugarcane growers, who receive the major part of the import rights dispensed by the Sugar Regulatory Administration. 15. During this period, annual imports of live cattle averaged more than 200,000 head. The num- ber declined significantly thereafter, and the cattle fattening business is now limited to Del Monte and San Miguel Corporation­Monterey Foods, which use by-products of their other businesses, that is, pineapple canning and beer manufacturing, respectively, as main cattle feed ingredients. Previously, live cattle were mostly imported for almost immediate slaughter, and the government's policy was thus aimed at promoting the slaughtering business rather than cattle production. References Aldaba, R. M. 2005a. "Policy Reversals, Lobby Groups and Economic Distortions." Discussion Paper 2005­04, Philippine Institute for Development Studies, Makati City, the Philippines. ------. 2005b. "The Impact of Market Reforms on Competition, Structure and Performance of the Philippine Economy." Discussion Paper 2005­24, Philippine Institute for Development Studies, Makati City, the Philippines. The Philippines 253 Anderson, K. 1995. "Lobbying Incentives and the Pattern of Protection in Rich and Poor Countries." Economic Development and Cultural Change 43 (2): 401­23. Anderson, K., and Y. Hayami, eds. 1986. The Political Economy of Agricultural Protection: East Asia in International Perspective. London: Allen and Unwin. Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Balisacan, A. M., N. Fuwa, and M. H. Debuque. 2004. "The Political Economy of Philippine Rural Development since the 1960s." In Rural Development and Agricultural Growth in Indonesia, the Philippines, and Thailand, ed. T. Akiyama and D. F. Larson, 214­93. Canberra: Asia Pacific Press. Bautista, R. M., and G. R. Tecson. 2003. "International Dimensions." In The Philippine Economy: Development, Policies, and Challenges, ed. A. M. Balisacan and H. Hill, 136­71. New York: Oxford University Press. Borrell, B., D. Quirke, B. de la Pena, and L. Noveno. 1994. Philippine Sugar: An Industry Finding Its Feet. Canberra: Center for International Economics. Clarete, R. L. 1991. "E.O. 470: The Economic Effects of the 1991 Tariff Policy Reform." Unpublished report, United States Agency for International Development­Philippines, Manila. Clarete, R. L., and J. A. Roumasset. 1990. "The Relative Welfare Cost of Industrial and Agricultural Policy Distortions: A Philippine Illustration." Oxford Economic Papers 42 (2): 462­72. Cororaton, C. B. 2000. "Philippine Tariff Reform: A CGE Analysis." Discussion Paper 2000­35, Philippine Institute for Development Studies, Makati City, the Philippines. David, C. C. 1983. "Economic Policies and Philippine Agriculture." Working Paper 83­02, Philippine Institute for Development Studies, Makati City, the Philippines. ------. 1994. "GATT-UR and Philippine Agriculture: Facts and Fallacies." Journal of Philippine Devel- opment 21 (1­2): 141­78. ------. 2003. "Agriculture." In The Philippine Economy: Development, Policies, and Challenges, ed. A. M. Balisacan and H. Hill, 175­218. New York: Oxford University Press. David, C. C., P. Intal, and A. M. Balisacan. 2007. "Distortions to Agricultural Incentives in the Philip- pines." Agricultural Distortions Working Paper 28, World Bank, Washington, DC. FAOSTAT Database. Food and Agriculture Organization of the United Nations. http://faostat.fao.org/ default.aspx (accessed May 2008). GTAP Database (Global Trade Analysis Project Database). Center for Global Trade Analysis, Depart- ment of Agricultural Economics, Purdue University. https://www.gtap.agecon.purdue.edu/ databases/v6/default.asp (version 6). Habito, C. F., and C. B. Cororaton. 2000. "WTO and the Philippine Economy: An Empirical and Analytical Assessment of Post-WTO Trade Reforms in the Philippines." Unpublished report, Agile Program, United States Agency for International Development­Philippines, Manila. Hara, Y. 1994. Tonan ajia shokoku no keizai hatten (The Economic Development of Southeast Asian Countries). Tokyo: Institute of Oriental Culture, University of Tokyo. Intal, P. S., and J. H. Power. 1991. "The Philippines." In Asia. Vol. 2 of The Political Economy of Agricul- tural Pricing Policy, ed. A. O. Krueger, M. Schiff, and A. Valdés, 149­94. World Bank Comparative Study. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Lindert, P. 1991. "Historical Patterns of Agricultural Policy." In Agriculture and the State: Growth, Employment, and Poverty in Developing Countries, ed. C. P. Timmer, 29­83. Ithaca, NY: Cornell University Press. Manasan, R. G., and V. S. Pineda. 1999. "Assessment of Philippine Tariff Reform: A 1998 Update." Unpublished working paper, Philippine Institute for Development Studies, Makati City, the Philippines. Pasadilla, G. O. 2006. "Preferential Trading Agreements and Agricultural Liberalization in East and Southeast Asia." Working Paper 11, Asia-Pacific Research and Training Network on Trade, Trade and Investment Division, United Nations Economic and Social Commission for Asia and the Pacific, Bogor, Indonesia. 254 Distortions to Agricultural Incentives in Asia Pimentel, A. 2006. "A Study on the Impact of the Philippine Tariff Reform Program: An Input-Output Model." Unpublished working paper, Southeast Asia Research Center for Agriculture, Los Banos, the Philippines. Power, J. H. 1971. "The Structure of Protection in the Philippines." In The Structure of Protection in Developing Countries, ed. B. Balassa, 271­80. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Tan, E. S. 1994. "Trade Policy Reforms in the 1990s: Effects of EO 470 and the Import Liberalization Program." Research Paper 94­11, Philippine Institute for Development Studies, Makati City, the Philippines. Tan, N. A. 1979. "The Structure of Protection and Resource Flows in the Philippines." In Industrial Promotion Policies in the Philippines, ed. R. M. Bautista and J. H. Power, 122­67. Makati City, the Philippines: Philippine Institute for Development Studies. Tyers, R., and K. Anderson. 1992. Disarray in World Food Markets: A Quantitative Assessment. New York: Cambridge University Press. World Development Indicators Database. World Bank. http://go.worldbank.org/B53SONGPA0 (accessed May 2008). 7 THAILAND Peter Warr and Archanun Kohpaiboon Thailand is a major net agricultural exporter, and its agricultural trade policy is dominated by this fact. The list of agricultural exports includes many of the most important agricultural products produced and consumed within the country, including the staple food, rice (exports of which account for between 30 and 50 percent of the total output of the product), and also cassava, sugar, rubber, and poultry products. The list of imported agricultural commodities is much shorter. Maize has been a net export in most years, but it was also a net import during some years in the 1990s. Soybeans were a net export for several decades, but, since the early 1990s, have been a net import. Palm oil has fluctuated between a net import and a net export, but, since the late 1990s, it has been a net export. Historically, Thailand's large agricultural surplus has always led to a degree of policy complacency regarding the agricultural sector. Agricultural importing countries are typically concerned about food security and raising agricultural productivity to reduce import dependence. In Thailand, these matters have not been a significant concern, although stabilizing food prices for consumers has been a recurrent theme of agricultural pricing policy. Until the 1980s, agricultural exports were viewed as a source of government revenue. Unlike manufacturing, traditional agriculture was not seen as a dynamic sector of the economy that might contribute to rapid growth. Because the price elasticity of supply for most agricultural products was low, at least in the short run, it was felt that the The authors gratefully acknowledge the excellent research assistance of Arief Ramayandi and the help- ful comments and assistance with data of the following colleagues: Ammar Siamwalla, Chalongphob Sussangkarn, and Wisarn Pupphavesa of the Thailand Development Research Institute; Isra Sarntisart of Chulalongkorn University; and Nipon Poapongsakorn, Prayong Netayarak, and Somboon Siriprachai of Thammasat University. They are also grateful for the helpful comments of workshop par- ticipants and the intensive assistance with spreadsheets by Marianne Kurzweil and Ernesto Valenzuela. 255 256 Distortions to Agricultural Incentives in Asia production of these products could be taxed heavily without creating a significant contraction in output. Moreover, most agricultural producers were impoverished, poorly educated, and politically unorganized. This was particularly the case among rice producers; so, taxing agriculture, especially rice, was politically attractive, and rice exports were taxed until 1986 (Siamwalla, Setboonsarng, and Patamasiriwat 1993). Because of greatly increased income per person, rapid urbanization, and the swing toward more democratic political institutions, policy has shifted from tax- ing agriculture to more neutral trade policies. This change has almost certainly owed more to politics--the political necessity of finding ways to attract the sup- port of the huge rural electorate and the desire of the urban electorate to establish better economic conditions for the farm population--than to a desire to liberalize agricultural trade for the efficiency reasons that economists emphasize. However, the transition away from taxing agriculture has not progressed much in the direc- tion of subsidizing agriculture. The key reason is that many important agricul- tural commodities are net export items, and this has made subsidizing agriculture problematic. Had the commodities farmers produce been competing with imports and thus amenable to the imposition of tariffs, there might have been more political pressure to protect farmers. Thailand is an active member of the Cairns Group of agricultural exporting countries within the membership of the World Trade Organization. Nonetheless, while its agricultural trade is relatively liberal, the country cannot be described as a free trader with regard to agricultural commodities. Within Thailand, opposi- tion to agricultural import liberalization is strong in the case of soybeans, palm oil, rubber, rice, and sugar. The protective measures include nontariff instruments that permit substantial discretion by government officials. The import controls include import prohibitions, strict licensing arrangements, local content rules, and a requirement for special case-by-case approval of imports. The commodities for which these restrictions are applied include the five above, plus onions, garlic, potatoes, pepper, tea, raw silk, maize, coconut products, and coffee. The inclusion of rice in this list of protected commodities may seem strange. Thailand is the world's largest exporter of rice and is undoubtedly one of the world's most efficient rice producers. Why should the rice industry require pro- tection from imports? Imports of rice are, in fact, prohibited unless they are specif- ically approved by the Ministry of Commerce. The Ministry of Agriculture and Cooperatives vigorously opposes any liberalization commitments with regard to rice. The reasons apparently relate to the ministry's wish to keep its options open with respect to rice policy in the event market conditions should change unex- pectedly. Sudden changes in the price of rice can have far-reaching political conse- quences. The domestic rice market operates almost entirely without government intervention, but the instruments for intervention are ever ready. Thailand 257 A lesser reason for the import controls on rice is that, as with most agricultural commodities, rice is a highly differentiated commodity. Not all grades of rice are produced efficiently within Thailand, and the government wishes to protect domestic producers from imports of the grades of rice that are closer substitutes for local grades on the consumption side than they are on the production side. The lower grades of rice produced in Vietnam, but not in Thailand, are an impor- tant example. Thailand's general exclusion list applies to agreements in the free trade area of the Association of Southeast Asian Nations and includes several agricultural industries, among which are rice, sugar, and palm oil (both crude and refined). Within government circles, discussions about problems in the agricultural sector revolve overwhelmingly around the treatment of Thai exports by other countries. Thailand's own agricultural import policy is a closed question. Problems have been encountered with a number of trading partners about environmental issues and sanitary and phytosanitary issues with Thailand's agricultural exports. These problems have included a well-publicized dispute with the United States over shrimps (environmental issues) and another with Australia over precooked frozen chicken (sanitary and phytosanitary issues). Poverty in the country is heavily concentrated in rural areas, and public opin- ion favors government support for the rural poor. Since the economic crisis of 1997­98 and especially during the government of Prime Minister Thaksin Shinawatra in 2001­06, a wide range of income support programs, cash grants to villages, and subsidized credit schemes have been introduced. Support for these schemes was a significant component of the populist economic policy agenda of the Thaksin government. However, the operations of few of these schemes, if any, depended on the prices faced by agricultural producers. Because they were not linked directly to the production of agricultural commodities, it seems that they were not distorting resource allocations. The results of our study make an assess- ment possible of whether the price incentives facing agricultural producers were, indeed, distorted relative to international prices during the period of populist government. The following section of the chapter briefly describes the changing structure of the Thai economy, especially the agricultural sector. The core of the analysis, con- tained in the section thereafter, relies on comparisons between the domestic prices and the international prices for major agricultural commodities and fertilizer and relates the price comparisons to tariff and nontariff barriers on the same prod- ucts. It focuses on whether the relative prices for traded commodities at wholesale have differed from the relative border prices, adjusted for transport and handling costs. The subsequent section extends this analysis to the farm level. The raw com- modities produced by farmers generally do not enter international trade directly. These raw commodities are inputs into the production of the processed 258 Distortions to Agricultural Incentives in Asia commodities that actually enter international trade. For example, rice produced at the farm level (paddy) must be milled before it can be traded internationally. Rice milling, transport, packaging, and storage are all costly activities, and several steps in the marketing chain intervene between farmers and the international market. This raises the controversial issue of the effect exerted on the prices actually received by farmers for their output, such as paddy, by the protection of processed commodities, such as milled rice, that is observed at the wholesale level and cap- tured by the price comparisons. In the penultimate section, we analyze this issue econometrically using price data, and, from this analysis, we derive the imputed rates of protection for farm-produced commodities. The final section concludes with a discussion of the future prospects for agricultural trade policy in Thailand. Economic Growth and Structural Change Over almost four decades, from 1968 to 2005, Thailand's economic output grew in real terms at an average annual rate of 6.5 percent. The broad characteristics of this growth are summarized in table 7.1. For ease of comparison with other Asian economies, the table distinguishes between the two-decade preboom period ending in 1986 and the following boom decade that preceded the Asian crisis of 1997­98. As the table shows, the growth rate of gross domestic product (GDP) during the boom decade was 9.5 percent, which was the most rapid GDP growth in the world and almost half again as rapid as the GDP growth during the two pre- boom decades. Output contracted during the crisis years of 1997­98. During the subsequent recovery, growth averaged a moderate 5.1 percent. In Thailand, as in many other rapidly growing economies, agricultural output increased more slowly than GDP, implying that agriculture had a declining share in aggregate output. The agricultural sector accounted for 32 percent of GDP in 1965. By 2004, this share had declined to 10 percent. Over the same period, the GDP share of industry rose from 23 to 43 percent, and the share of services Table 7.1. Real GDP Growth and Its Sectoral Components, Thailand, 1968­2005 (average annual percent) Pre-boom, Boom, Crisis, Recovery, All years, Sector 1968­86 1987­96 1997­99 2000­05 1968­2005 Total 6.7 9.5 2.5 5.1 6.5 Agriculture 4.5 2.6 0.1 3.6 3.5 Industry 8.5 12.8 1.7 6.3 8.5 Services 6.8 9.0 3.6 4.2 6.2 Source: Author calculations based on data in World Development Indicators Database 2008. Thailand 259 Table 7.2. Subsector Shares of Agricultural Value Added, Thailand, 1975­2000 (percent) Subsector 1975 1980 1985 1990 1995 2000 Paddy 38.0 30.3 34.7 24.9 26.9 26.1 Maize 6.4 4.3 4.2 3.7 3.7 3.4 Other cereals 0.5 0.6 0.5 0.2 0.1 0.2 Cassava 4.2 7.6 5.5 6.6 5.2 2.5 Beans and nuts 2.4 2.5 3.7 3.0 2.1 1.7 Vegetables 11.7 10.4 9.1 12.7 9.9 10.6 Fruits 11.4 15.0 10.5 10.9 11.1 15.8 Sugarcane 5.9 5.4 3.2 6.7 5.2 5.3 Coconuts 1.4 1.7 1.8 1.2 0.9 0.7 Palm nuts and palm oil 0.0 0.1 0.6 1.2 1.2 1.4 Rubber 2.2 4.6 8.4 10.2 17.5 12.4 Other crops 5.7 5.2 5.3 4.3 4.3 4.3 Cattle and buffalo 2.5 3.3 5.3 6.3 3.9 4.8 Swine 3.2 3.0 1.6 1.9 1.7 1.5 Poultry 1.1 2.0 4.0 3.6 3.9 6.6 Other livestock 3.6 4.0 1.9 2.7 2.1 2.9 Total 100 100 100 100 100 100 Source: Author calculations based on data in NESDB, various. grew from 45 to 47 percent. Part of this long-term contraction is explained by the declining terms of trade for the country's agricultural exports (Warr and Kohpaiboon 2007). For a more detailed study of the changing composition of the agricultural sec- tor, one may use input-output tables, which are available at five-year intervals from 1975 to 2000 (NESDB, various). Over this period, the value added in paddy production (unmilled rice produced at the farm level) declined from 38 to 26 per- cent of total agricultural value added (table 7.2). Changes in the distribution of expenditure as incomes increased explain most of this change. As incomes rise, the expenditure on starchy staples typically declines as a share of total expendi- tures. The shares of maize and cassava similarly declined, but the shares of fruit, poultry, cattle, and rubber rose. For almost all agricultural commodities, the share of intermediate inputs in the value of total output increased significantly over 1975­2000 (Warr and Kohpaiboon 2007). In paddy production, for example, the share rose from 14 to 30 percent. For the entire agricultural sector, this cost share rose from 21 to 37 percent over the same period. Most intermediate goods used in agriculture are 260 Distortions to Agricultural Incentives in Asia domestically produced, but the share of imports in total intermediate input use increased from 10 to 17 percent (Warr and Kohpaiboon 2007). There have been substantial changes in the pattern of sales among agricultural products. In 1975, sales of agricultural products to intermediate users (millers and processors) accounted for 57 percent of total sales, but, by 2000, these sales had risen to 70 percent. Almost all paddy is milled into edible rice commercially rather than on-farm. Paddy is not exported or imported, but milled rice has his- torically been an important export item, as has refined sugar. Cassava is similarly exported in the form of processed animal feed. Rubber exports have become increasingly significant since the 1990s. Soybeans have become a major net import and are used for processed foods and for animal feed (Warr and Kohpaiboon 2007).1 The Changing Structure of Assistance at the Wholesale Level In their definitive studies of agricultural price policy in Thailand up to the mid- 1980s, Siamwalla and Setboonsarng (1989, 1991) make the point that policies on the various agricultural commodities were determined individually in response to circumstances that varied depending on the commodity rather than forming part of a single, integrated agricultural policy approach. For this reason, they argue, it is best to consider the main commodities one at a time, which they then do for rice, sugar, maize, and rubber. The discussion that follows will also rely on this strategy, except that the range of commodities includes cassava, soybeans, and palm oil, in addition to the four reviewed by Siamwalla and Setboonsarng. Our analysis also considers a major input, urea fertilizer. Following this commodity- specific review, we address common themes, if any, that may run through agricul- tural policy as a whole. The main focus of our study's methodology is government-imposed distor- tions that create a gap between domestic prices and the prices that would emerge under free market conditions. (For a detailed description of the methodology, see Anderson et al. 2008 and appendix A.) Since it is not possible to understand the characteristics of agricultural development through a sectoral view alone, the methodology not only estimates the effects of direct agricultural policy measures, but, for comparative evaluation, it also includes estimates of distortions in non- agricultural tradable sectors. Specifically, we compute nominal rates of assistance (NRAs) for farmers, including an adjustment for direct interventions on tradable inputs such as fertilizer.2 We also calculate NRAs for nonagricultural tradables for use with the NRAs for agricultural tradables to generate relative rates of assistance (RRAs). Thailand 261 We conduct our analysis at the wholesale level; so, in what follows, the domestic price means the domestic wholesale price. In the calculation of the NRAs, the bor- der prices are amended by the transport and handling costs involved in adjusting the price of imports from the cost, insurance, and freight level to the domestic wholesale level and in adjusting the price of exports from the domestic wholesale level to the free on board level. (The relevant transport and handling costs are sum- marized in Warr and Kohpaiboon 2007.) These adjustments are required to obtain prices that are comparable with domestic wholesale prices. The border prices adjusted for transport and handling costs may then be interpreted as an indication of domestic wholesale prices as they would be in the absence of protection. Rice From the end of World War II to 1986, the Thai government taxed rice exports. There were four individual instruments of export taxation, each with distinct legal foundations, and each under the control of different parts of the government bureaucracy. The revenues these various instruments generated were directed toward separate parts of the government. Siamwalla and Setboonsarng (1989, 1991) describe these independent tax streams and point out that their combined effect was a rate of export taxation of around 40 percent from the late 1950s to the early 1970s. The rate increased to around 60 percent during the commodity price boom of 1972­74, but subsequently diminished quickly to about 20 percent. There was another peak, at about 40 percent, at the time of the second Organiza- tion of the Petroleum Exporting Countries oil price shock in 1979­80 and then a steady decline until all four forms of tax were suspended in 1986. Rice exports have remained untaxed for the two decades since then.3 The implications of these events for actual prices are summarized in figure 7.1, which is based on the data assembled in Warr and Kohpaiboon (2007). As with each similar figure presented thereafter on another agricultural commodity, figure 7.1 compares domestic wholesale prices with border prices for rice of com- parable quality. Since rice is a net export, border price in the figure means export price, adjusted for transport and handling costs between the wholesale level and the export level. The calculations of the nominal rates of protection (NRPs) that emerge are similar to those that would be inferred from the rates of taxation described above, except that the NRPs after 1986 are not zero, but average around 6 percent. It is possible that the transport and handling costs between the wholesale and free on board locations are not fully accounted for in the data in Warr and Kohpaiboon (2007). For rice, the data in figure 7.1 support the view that the domestic market has received zero protection and zero subsidies; moreover, rice is also no longer taxed. 262 Distortions to Agricultural Incentives in Asia Figure 7.1. Price Comparisons and NRPs at Wholesale, Rice, Thailand, 1968­2005 14,000 0 12,000 10 ton 10,000 20 metric NRP 8,000 30 per ,% 6,000 40 baht 4,000 50 price, real 2,000 60 0 70 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 year left axis: border price right axis: NRP, % domestic price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. Maize Maize was a net Thai export item until the 1990s. In 1992 and 1995­2000, imports dominated. Maize subsequently reverted and is now a net export good once more. In 1965­81, the government intervened in the export market to preserve the exports to Japan and Taiwan, China, where they were used primarily for animal feed. In both markets, season-long stability in supply was an import requirement. To ensure the fulfillment of the domestic contracts needed to ensure this stability, the government imposed quota restrictions on exports to markets other than the markets of these two economies. The effects of this policy included an increase in the price volatility passed on to domestic producers and average producer earn- ings that were somewhat reduced. Meanwhile, Malaysia, Singapore, and other countries closer to Thailand had developed their own livestock industries. The Japanese and Taiwan, China markets therefore seemed less crucial, and, by 1981, the export controls on maize were removed. The data in figure 7.2 indicate that there was roughly zero protection for the maize industry. Whether maize was a net import product or a net export product, this outcome does not seem to have varied in any systematic way. Thailand 263 Figure 7.2. Price Comparisons and NRPs at Wholesale, Maize, Thailand, 1968­2005 6,000 20 10 ton 5,000 0 metric 4,000 NRP 10 per 3,000 ,% 20 baht 2,000 30 price, 1,000 real 40 0 50 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 year left axis: border price right axis: NRP, % domestic price farm price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. Cassava Thailand's cassava exports developed for the supply of animal feed to European and some Asian markets, including Taiwan, China. Because of the quota restric- tions of the European Union, rents were attached to export quotas from Thailand, which, in turn, led to corruption in the allocation of the quotas. The rents associ- ated with the quotas are analogous to a privately collected export tax. As a result of the rents, the export price exceeded the domestic price by amounts averaging around 10 percent (figure 7.3). Soybeans Soybeans were a net export item from 1960 until 1988. By 1992, they were a net import item. During the export period, exports were taxed, but, beginning in 1995, the trade regime shifted nominally to tariff quotas. The operation of the tariff quotas is described in Warr and Kohpaiboon (2007). Within the import quota volume permitted, soybeans could be imported at low or zero tariffs. Beyond the quota, the applied tariff was set at the maximum amount permitted by World Trade Organization obligations, which varied between 80 and 90 percent. 264 Distortions to Agricultural Incentives in Asia Figure 7.3. Price Comparisons and NRPs at Wholesale, Cassava, Thailand, 1969­2004 4,500 10 4,000 5 ton 3,500 0 metric 3,000 5 NRP per 2,500 10 ,% 2,000 baht 15 1,500 price, 1,000 20 real 500 25 0 30 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 year left axis: border price right axis: NRP, % domestic price farm price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. Figure 7.4 indicates that the transition of soybeans from a net export to a net import in 1992 coincided with a shift from negative NRPs of around 20 percent to positive NRPs of 30­40 percent. Sugar In discussions on agricultural economics, the sugar industry is often considered an outlier in the treatment it receives from trade policy. Thailand is no exception. Sugar was an imported item until the late 1950s, but has been a net export item since then. Nonetheless, it receives protection through a home price scheme. This type of scheme involves taxing consumers and using the proceeds to subsidize exports. A scheme of this sort was practiced in the Australian sugar and dairy industries in the 1950s and 1960s (Sieper 1982). It is said that a Thai economics student at an Australian university learned about the scheme and brought the idea to Thailand, where the scheme was applied in the sugar industry long after it had been abandoned in Australia. A home price scheme drives up the domestic consumer and producer prices of the product to subsidize producers at the expense of consumers. The scheme Thailand 265 Figure 7.4. Price Comparisons and NRPs at Wholesale, Soybeans, Thailand, 1984­2005 16,000 60 14,000 ton 40 12,000 metric 20 10,000 NRP per 8,000 0 ,% baht 6,000 20 price, 4,000 40 real 2,000 0 60 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 year left axis: border price right axis: NRP, % domestic price farm price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. requires that leakage be prevented from the export market to the more profitable domestic market. In most industries, this is difficult. Reimporting for domestic consumption must also be restricted, and, as Corden (1974) points out, this may be achieved through a restrictive tariff. Because the scheme is self-financing, it is attractive for finance ministries. However, the scheme is a protectionist device, and has a notable limitation: the capacity of the consumption tax to subsidize exports is reduced if the volume of exports comes to represent a substantial share of total output (exports, plus domestic consumption). This has been an issue in the case of the sugar industry in Thailand. Siamwalla and Setboonsarng (1989, 1991) attribute the political power of the Thai sugar industry to technological changes in milling that call for large mills and the precise scheduling of sugar deliveries to the mills. Milling is highly capital intensive, and, during the sugar processing season, processing plants must be fully utilized. Growers and millers have bickered over prices, but they have been able to combine their efforts to lobby the government for intervention on their behalf; other agricultural export industries in Thailand have been unable to achieve this success. Sugar growers and millers are highly organized. The technological 266 Distortions to Agricultural Incentives in Asia Figure 7.5. Ratios, Consumer Prices to Border Prices and Miller Prices to Grower Prices, Sugar, Thailand, 1968­2005 3.5 1.8 1.6 3.0 miller price 1.4 2.5 1.2 price/grower 2.0 1.0 price/border 1.5 0.8 0.6 price 1.0 0.4 consumer 0.5 0.2 0 0 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 year left axis: consumer price/border price right axis: miller price/grower price Source: Warr and Kohpaiboon 2007. changes in sugar milling have also helped restrict leakage from the export market to home consumption because the mills are large and few in number. The scheme has stabilized consumer prices relative to export prices. Figure 7.5 shows two series: the ratio of consumer prices to border prices (left axis) and the ratio of miller prices to grower prices (right axis). The peak export prices of the early 1970s were not transmitted to consumers or producers, and the NRPs for sugar (calculated as the percent deviation of the grower price from the export price) were negative at the time (figure 7.6). However, during most of the period of the operation of the scheme, consumer and producer prices were well above export prices. Since the mid-1980s, the NRPs have averaged over 60 percent. Although it is exported, sugar is the most heavily protected agricultural subsector by far. Palm oil Palm oil has sometimes been a net import item and sometimes a net export item. Although the palm oil industry has been a net exporter since 1998, a system of import quotas remains in place (see Warr and Kohpaiboon 2007). Figure 7.7 shows the NRPs for palm oil at wholesale. Figure 7.6. Price Comparisons and NRPs at Wholesale, Sugar, Thailand, 1968­2005 25,000 120 100 ton 20,000 80 metric 60 15,000 NRP per 40 ,% 20 baht 10,000 0 price, 5,000 20 real 40 0 60 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 year left axis: border price right axis: NRP, % grower price miller price consumer price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. Figure 7.7. Price Comparisons and NRPs at Wholesale, Palm Oil, Thailand, 1995­2004 20,000 100 90 ton 16,000 80 70 metric 12,000 60 NRP per 50 ,% baht 8,000 40 30 price, 4,000 20 real 10 0 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 year left axis: border price right axis: NRP, % domestic price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. 268 Distortions to Agricultural Incentives in Asia Figure 7.8. Price Comparisons and NRPs at Wholesale, Rubber, Thailand, 1968­2005 40,000 15 35,000 10 ton 5 30,000 0 metric 25,000 5 NRP per 20,000 10 ,% baht 15,000 15 20 price, 10,000 25 real 5,000 30 0 35 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 year left axis: border price right axis: NRP, % domestic price farm price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. Rubber Rubber is a net export item, and the rubber industry was subject to an export tax for many years. The manner of calculating the tax meant that the rate drifted upwards along with inflation, and, because of the inflation in the 1970s, the export tax rate on rubber had reached 26 percent by the early 1980s. Pressure from mem- bers of Parliament representing rubber production areas in the southern parts of the country led to a revision of the method of calculating the tax. This generated a return to the lower tax rates that had been in place in the 1960s. Figure 7.8 con- firms that, since 1990, the tax rate on rubber has been close to zero. Fertilizer Urea is imported for use as fertilizer, the rates of tariff protection for these imports have been declining. The import taxation implies that there is disprotection (tax- ation) for the agricultural industries using this input. The decline in tariff rates began in the early 1990s, and, by the early 2000s, the rates were negligible. This policy change is confirmed by the price comparisons reported in figure 7.9. NRPs Thailand 269 Figure 7.9. Price Comparisons and NRPs at Wholesale, Urea Fertilizer, Thailand, 1984­2005 8,000 30 7,000 25 ton 6,000 20 metric 5,000 NRP 15 per 4,000 ,% 10 baht 3,000 5 price, 2,000 real 1,000 0 0 5 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 year left axis: retail price right axis: NRP, % wholesale price border price Source: Warr and Kohpaiboon 2007. Note: NRP 100*(domestic price border price) border price. have declined steadily and are currently close to zero. This treatment of fertilizer--steadily declining rates of taxation--contrasts with the treatment in several neighboring countries, where fertilizer use has tended to be subsidized as part of a general program of agricultural subsidization. Imputed Assistance at the Farm Level Our discussion of protection rates has focused so far on the effects that policy interventions have on the wholesale market. In this section, we extend the analysis to consider the way protection (or its opposite) in the wholesale market produces price effects at the farm level. Theory One of the aims of agricultural protection policy is to influence prices at the farm- gate. However, the goods produced by farmers seldom enter international trade directly. The raw commodities produced by farmers are generally nontraded, 270 Distortions to Agricultural Incentives in Asia whereas the commodities that enter international trade are the processed or par- tially processed versions of these raw products. Between the nontraded raw prod- uct produced by the farmer and the traded processed commodity that enters international trade, there may be several other steps, including transport, storage, milling, processing, and repackaging. However, border protection policy operates on the goods that enter interna- tional trade either as exports or as imports. It does not operate on the raw com- modities produced by farmers. Protection at the farm level is therefore a derived effect. It depends on the extent to which the policies applied to the trade in processed agricultural goods induce changes in the prices of these goods and then transmit the changes to the prices faced by farmers. An issue thus arises about the degree to which price changes at wholesale that are induced by protection policy affect the prices received by farmers for the raw products they sell. We construct a simple econometric model to investigate this issue. We use the notational convention that uppercase roman letters (such as X) denote the level of the values of variables and that lowercase roman letters (such as x) denote the nat- ural logarithms of these values. Thus, x ln X. Protection at the wholesale level is therefore defined as: PW it P*it(1 TW), it (7.1) where PW denotes the level of the wholesale price of commodity i at time t; P*it is it the corresponding border price expressed in the domestic currency and adjusted for the handling costs involved in transferring the commodity from the cost, insurance, and freight price level to the domestic wholesale price level in the case of an import or in transferring the commodity from the wholesale price level to the free on board price level in the case of an export. The NRP at the wholesale level is given by Tit . In this discussion, both the border price and the NRP are W treated as exogenous variables. The border price is determined by world markets, and we presume that the country under discussion is a price taker. The NRP is determined by the government's protection policy. The farmgate price of the raw material is denoted by PFit, and its logarithm, pFit, is related to the logarithm of the wholesale price as follows: pFit ai bipW it uit, (7.2) where ai and bi are coefficients, and uit is a random error term. The coefficient bi is the pass-through or transmission elasticity. The estimated values of the coeffi- cients ai and bi are denoted by a^i and b^i, respectively. The econometric estimation of these parameters is discussed below. The estimated coefficients are used as follows. We estimate the logarithm of the farm price that would obtain in the absence of any protection as: p^it F* a^i b^ipW*, it (7.3) Thailand 271 where pit W* is the estimated value of the wholesale price that would obtain in the absence of protection so that pit W* ln Pit . This is then compared with the esti- W* mated value of the wholesale price in the presence of protection, as follows: p^it F a^i b^ipW. it (7.4) We denote the antilogs of p^it and p^it by P^it and P^ it , respectively, and then esti- F F* F F* mate the NRP at the farm level as: T^it = (P^it F F P^ it ) P^it. F* F (7.5) It is important to observe that the value of the protection-inclusive farmgate price we use in these calculations is the price we have estimated using the econo- metric model (equation 7.4) rather than the actual price given by the raw data. We do this because our intention is to use the model to estimate the change in the farmgate price generated by the protection at the wholesale level. Thus, both the protection-inclusive and the protection-exclusive prices used in (7.5) are the pre- dicted values obtained through the model. The implied NRP at the farmgate can be related to the NRP at wholesale, as fol- lows. Substituting P^it F A^i(PW)b^i and P^it F* it A^i(PW*)b^i into equation (7.5), where A^i it is the antilog of a^i, rearranging, and using equation (7.1), we obtain the simple expression: T^it F (1 TW)b^i it 1. (7.6) Obviously, if TW it 0, then T^it F 0, regardless of the value of b^i. Similarly, if b^i 0, then T^it F 0, regardless of the value of TW. Also, if b^i it 1, then T^it = TW. It F it may be readily seen that, if TW it 0, then T^F TW as b^i it 1 and T^F TW as b^i it 1. If TWit 0, then T^F TW as b^i it 1, but T^F TW as b^i it 1. Econometric application The purpose of the econometric analysis is to estimate the parameter b^i for each commodity.4 We conduct the analysis for each commodity by relying on time series price data in which each variable is expressed in logarithms, and each is deflated by the GDP deflator for Thailand: the farmgate price (LFP), the wholesale price (LWP), and the log of the international price, adjusted by the nominal exchange rate and transport and handling costs (LIP). We first test each of the series for the existence of a unit root. The null hypoth- esis of a unit root was rejected for all price series for all commodities except soy- beans. (One should recall that the price series, relying on the GDP deflator, are real, not nominal.) However, in the case of soybeans (the two price series for which the null hypothesis of a unit root could not be rejected), the series were not cointegrated. For all commodities except soybeans, the price series were thus con- sidered stationary. 272 Distortions to Agricultural Incentives in Asia First, we produced ordinary least squares estimates of equation (7.2). In most cases, autocorrelation was a problem, and an AR(1) correction term was included to eliminate it. The correction term accomplished this task effectively. The ordinary least squares estimates assume that LFP is endogenous and LWP is exogenous. These assumptions were tested using Hausman's endogeneity test. In the case of each commodity, the null hypothesis that LWP was (weakly) exoge- nous to LFP failed to be rejected, confirming the validity of the estimates. Reverse Hausman's tests were also conducted, and the null hypothesis that LFP was exoge- nous to LWP was rejected in every case. These results support the validity of the use of the ordinary least squares framework to estimate the transmission elasticity from LWP to LFP, treating LWP as exogenous. For completeness, instrumental variable estimates were produced for each commodity using LIP as the instru- ment for LWP. The resulting estimates of b^i differed from the ordinary least squares estimates (some larger, some smaller), but not by much. Table 7.3 summarizes the estimates. All the ordinary least squares estimates of transmission elasticity were significantly different from zero and showed the expected positive signs. This is an important point. It is often asserted that the commodity price changes arising at the wholesale level (because of protection or international price movements) are prevented by middlemen from being trans- mitted to farmers. This hypothesis is strongly rejected by the Thai data. The trans- mission elasticities are not zero. Economists often assume that the transmission elasticities are unity. However, the estimated values are generally less than unity; Table 7.3. Estimates of Transmission Elasticities from Wholesale to Farm Prices, Thailand Commodity Estimated elasticity t-statistic Rice 0.76 (7.30) Maize 0.81 (14.38) Cassava 1.07 (8.20) Soybeans 0.80 (11.23) Sugar 0.53 (3.93) Palm oila [0.90] (19.97) Rubber 0.90 (19.97) Fertilizer 0.89 (17.70) Source: Author calculations. Note: The table has been created using the data and methodology discussed in the text. The estimates shown relate to the parameter bi in equation (7.2). a. An estimate for palm oil has not been possible because of insufficient data points. The estimated value for rubber has been used instead. Thailand 273 most are between 0.7 and 0.9. In one case (sugar), the estimate is somewhat lower (0.53), and, in another (cassava), the estimated value slightly exceeds unity, but is not significantly different from unity.5 It is likely that the true transmission elasticities change over time, but the limited data available for this exercise have made it necessary to assume that the true values remain constant. Estimation of assistance at the farm level Given the estimated value of the transmission elasticity, we use equation (7.6), together with the estimated NRPs at the wholesale level discussed above, to pro- duce estimates of the imputed NRAs at the farm level. These are shown in Warr and Kohpaiboon (2007). Because the estimated values of the transmission elastic- ity are between zero and unity (except for the case of cassava), the imputed NRAs at the farmgate are somewhat lower in absolute value than the NRAs at the whole- sale level, but, because of the assumption of constant transmission elasticities over time, they closely track the pattern of the results at the wholesale level. Aggregate Measures of Agricultural Assistance In this section, we calculate aggregate measures of rates of assistance using the infor- mation assembled from the preceding analysis and following, as much as possible, the methodology outlined in Anderson et al. (2008) and appendix A. The annual calculations reported in this section fluctuate somewhat from year to year. Interna- tional and domestic price changes from year to year alter the protective effects of all instruments of protection except ad valorem tariffs. In addition, the time required for domestic prices to adjust to international price changes means that the annual data on price differences produce some spurious variations from one year to the next. Our interest is in broad trends rather than in these annual fluctuations. Table 7.4 reports estimates of the NRAs at the farmgate for all commodities. These estimates take account of the assistance for fertilizer inputs. The NRAs are calculated as the corresponding NRPs (discussed above), minus the product of the cost share of fertilizer in the production of the relevant commodity and the con- sumer tax equivalent of the import protection for the fertilizer industry. The con- sumer tax equivalent for fertilizer is positive every year but one, although the rates of taxation have declined since the mid-1980s. The NRAs for covered products at wholesale are therefore below the NRAs at the farmgate for every commodity for which fertilizer is an input in production. The broad pattern in the NRAs for farmers is otherwise similar to the pattern in the NRAs discussed above. The NRAs are negative in all years for rice and in most years for maize, cassava, and rubber. For these commodities, the absolute magnitudes of these negative 274 Table 7.4. NRAs for Covered Agricultural Products, Thailand, 1970­2004 (percent) Indicator, product 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Exportablesa 26.7 19.4 11.1 11.7 9.2 3.8 0.6 Soybeansb -- -- -- 19.9 n.a. n.a. n.a. Rice 30.1 28.3 18.0 15.0 16.2 11.0 7.6 Maize 2.2 2.6 2.2 7.6 4.5 11.5 0.3 Cassava 23.1 0.8 9.0 16.6 10.8 13.8 10.0 Sugar 12.6 3.2 12.7 36.8 34.0 22.4 12.6 Rubber 0.5 8.7 17.9 13.3 4.4 1.1 0.2 Poultry 32.9 16.1 26.8 7.1 11.0 17.8 20.4 Palm oil -- -- n.a. n.a. n.a. 12.6 18.3 Import-competing productsa 4.8 1.9 45.3 22.0 6.4 34.4 4.7 Soybeansb n.a. n.a. n.a. n.a. 27.5 21.5 30.0 Pig meat 4.8 1.9 51.7 20.8 1.5 36.5 1.8 Palm oil -- -- 25.7 32.2 26.5 n.a. n.a. All covered productsa 25.8 18.4 8.4 9.7 7.7 1.1 0.6 Dispersion, covered productsc 25.0 20.8 28.5 29.3 25.1 22.9 16.7 Coverage, at undistorted prices 65 65 68 71 71 75 78 Source: Warr and Kohpaiboon 2007. Note: -- no data are available. n.a. not applicable. a. Weighted averages; the weights are based on the unassisted value of production. b. For exportables, 1985­89 refers to 1984­91. For import-competing products, 1990­94 refers to 1992­94, and 2000­04 refers to 2000­03. c. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of covered products. Thailand 275 rates have declined over time. For soybeans, the nominal rate was negative until soybeans became a net import item in the early 1990s. Since then, soybeans have been significantly protected. Sugar has been a protected commodity in almost all years. The weighted average for all covered products was more than 25 percent before the later 1970s, but the mean rate of taxation has since fallen to virtually zero. The dispersion of rates for individual commodities around this mean has not fallen much though, suggesting that there is still considerable scope for policy reform to reduce distortions within the farm sector (bottom of table 7.4). We have calculated the RRAs in agriculture to take into account not only the NRAs in agriculture, but also the NRAs in manufacturing and other nonfarm tradable sectors. We estimate the average NRAs for import-competing manufac- turing based on data in Nicita and Olarreaga (2006), while we assume that the average NRAs for other nonfarm tradable sectors are zero.6 The weighted average of these two sets of NRAs is then calculated using weights from the input-output tables of the National Economic and Social Development Board (NESDB, vari- ous). The estimated RRAs are negative every year, but have declined in absolute value from more than 30 percent in the early 1970s to around 7 percent in recent years (table 7.5 and figure 7.10). These estimates suggest that, over the past four decades, agriculture has undergone a shift from a severely taxed sector to a mildly taxed sector (both net). Conclusions and Prospects for Future Reform As Thailand has industrialized, successive governments have become more inter- ested in intervening on behalf of agricultural producers. Nonetheless, because the country is a major agricultural exporter, the scope for a policy of protection as a means of influencing domestic commodity prices has been limited. This chapter has used comparisons of the prices of agricultural commodities in domestic mar- kets and in international markets as a means of examining the magnitude of the government's interventions. The direct taxation of agricultural exports has been gradually eliminated. This has been important in the case of rice, on which the high rates of export taxation before the mid-1980s have been abolished. Rubber exports, taxed prior to 1990, have been untaxed since then. Cassava exports have continued to be taxed to a minor extent through the system of export quotas. Fertilizer is a major input into agricultural production, and the effective taxation of fertilizer use has been steadily eliminated since the early 1990s. Maize exports have been consistently untaxed, as have chicken exports. Most of this evolution in taxation and protec- tion has involved eliminating the price distortions that once disfavored agricul- tural export industries. 276 Table 7.5. NRAs in Agriculture Relative to Nonagricultural Industries, Thailand, 1970­2004 (percent) Indicator 1970­74 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Covered productsa 25.8 18.4 8.4 9.7 7.7 1.1 0.6 Noncovered products 10.4 5.8 11.3 3.4 0.9 10.1 1.4 All agricultural productsa 20.3 14.0 2.0 6.2 5.7 1.7 0.2 Non-product-specific assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total agricultural NRAb 20.3 14.0 2.0 6.2 5.7 1.7 0.2 Trade bias indexc 0.18 0.20 0.37 0.24 0.14 0.27 0.03 NRA, agricultural tradables 23.1 15.9 2.3 6.9 6.4 1.8 0.2 NRA, nonagricultural tradables 16.1 16.0 14.2 11.1 10.0 8.9 7.8 RRAd 33.7 27.5 14.4 16.3 14.9 6.5 7.4 Source: Warr and Kohpaiboon 2007. a. Including product-specific input subsidies. b. Including product-specific input subsidies and non-product-specific assistance. Represents total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. c. The trade bias index is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. d. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. Thailand 277 Figure 7.10. NRAs for Agricultural and Nonagricultural Tradables and the RRA, Thailand, 1970­2004 30 20 10 % 0 RRA, 10 and 20 NRA 30 40 50 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 year NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Warr and Kohpaiboon 2007. Note: For the definition of the RRA, see table 7.5, note d. Three commodities depart from this general liberalization in agricultural mar- kets. Soybeans were an export before 1992 and have been a net import item since then. The imports have been subject to quota restrictions. The shift from a net export item to a net import item coincided with a switch from negative to positive NRPs. Since the early 1990s, the domestic soybean industry has benefited from an NRP of around 30­40 percent. Sugar is an export commodity, but the domestic sugar industry is protected by a system that taxes domestic consumers and trans- fers the revenue to producers. NRPs have averaged over 60 percent. The political power of the highly capital-intensive sugar milling industry is behind this pattern of protection. The case of palm oil is qualitatively similar to sugar, but the rates of protection are somewhat lower. Government interventions on behalf of rural people have been important, but they have generally not taken the form of interventions in agricultural commod- ity markets. Cash transfers to village organizations, subsidized loan schemes not linked to agricultural production, and a generally good system of public infra- structure have been the main instruments. The prospects for more trade liberal- ization are not encouraging unless this occurs through bilateral preferential trad- ing arrangements such as the scheme proposed with the United States.7 278 Distortions to Agricultural Incentives in Asia Notes 1. A full description of the trading position of the major agricultural commodities is provided in the appendix tables in Warr and Kohpaiboon (2007). 2. The price variables and the formula used in these NRA calculations are summarized in the appendix tables in Warr and Kohpaiboon (2007). Annual price data are also included in those tables. 3. The economic effects of the rice export taxes, including the distributional effects, are explored in Warr (2001). See also the analyses of Pinthong (1976, 1984); Wong (1978); Meenaphant (1981); Barker and Herdt (1985); Roumasset and Setboonsarng (1988); Somporn and Poapongsakorn (1995); Warr and Wollmer (1997); and Choeun, Godo, and Hayami (2006). 4. The results are summarized here. Details of the econometric analysis are available upon request. 5. There is no theoretical reason to suppose that the true value of the transmission elasticity is nec- essarily below unity. For example, if all margins between the farmgate and the wholesale level remain constant in nominal terms as the wholesale price changes, the percent change in the derived farmgate price would necessarily exceed the percent change in the wholesale price. The transmission elasticity would therefore exceed unity. 6. Because the Nicita and Olarreaga (2006) data are incomplete, we have assumed that the NRAs for manufacturing before 1982 and after 2002 are the same as the respective 1982 and 2002 NRAs cal- culated by Nicita and Olarreaga. This undoubtedly understates rates of manufacturing protection in the 1970s and overstates them post-2002. More complete estimates for manufacturing would therefore reinforce our broad conclusions rather than undermine them. 7. A bilateral trade arrangement with the United States was under negotiation up to February 2006, but the negotiations were suspended thereafter. The protection of the soybean industry was expected to be an important issue in these negotiations. References Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Barker, R., and R. W. Herdt. 1985. The Rice Economy of Asia. With Beth Rose. Washington, DC: Resources for the Future. Choeun, H., Y. Godo, and Y. Hayami. 2006. "The Economics and Politics of Rice Export Taxation in Thailand: A Historical Simulation Analysis, 1950­1985." Journal of Asian Economics 17 (1): 103­25. Corden, W. M. 1974. Trade Policy and Economic Welfare. Oxford: Clarendon Press. Meenaphant, S. 1981. "An Economic Analysis of Thailand's Rice Trade." PhD dissertation, Rice University, Houston. NESDB (National Economic and Social Development Board). various years. "Input-Output Table of Thailand." Spreadsheet, NESDB, Bangkok. Nicita, A., and M. Olarreaga. 2006. "Trade, Production, and Protection, 1976­2004." World Bank Economic Review 21 (1): 165­71. Pinthong, C. 1976. "A Price Analysis of the Thai Rice Marketing System." PhD dissertation, Stanford University, Palo Alto, CA. ------. 1984. "Distribution of Benefit of Government Rice Procurement Policy in Thailand." [in Thai] Thammasat University Journal 13 (2): 166­87. Roumasset, J. A., and S. Setboonsarng. 1988. "Second-Best Agricultural Policy: Getting the Price of Thai Rice Right." Journal of Development Economics 28 (3): 323­40. Siamwalla, A., and S. Setboonsarng. 1989. Trade, Exchange Rate, and Agricultural Pricing Policies in Thailand. Washington, DC: World Bank. ------. 1991. "Thailand." In Asia. Vol. 2 of The Political Economy of Agricultural Pricing Policy, ed. A. O. Krueger, M. Schiff, and A. Valdés, 236­80. World Bank Comparative Study. Baltimore: Johns Hopkins University Press; Washington, DC: World Bank. Thailand 279 Siamwalla, A., S. Setboonsarng, and D. Patamasiriwat. 1993. "Agriculture." In The Thai Economy in Transition, ed. P. G. Warr, 81­117. Cambridge: Cambridge University Press. Sieper, E. 1982. Rationalizing Rustic Regulation. Sydney: Center for Independent Studies. Somporn, I., and N. Poapongsakorn. 1995. "Rice Supply and Demand in Thailand: The Future Outlook." Paper presented to the Sectoral Economic Program, Thailand Development Research Institute, Bangkok. Warr, P. G. 2001. "Welfare Effects of an Export Tax: Thailand's Rice Premium." American Journal of Agricultural Economics 83 (4): 903­20. Warr, P. G., and A. Kohpaiboon. 2007. "Distortions to Agricultural Incentives in Thailand." Agricultural Distortions Working Paper 25, World Bank, Washington, DC. Warr, P. G., and F. Wollmer. 1997. "Testing the Small Country Assumption: Thailand's Rice Exports." Journal of the Asia Pacific Economy 2 (2): 133­43. Wong, C. M. 1978. "A Model for Evaluating the Effects of Thai Government Taxation of Rice Exports on Trade and Welfare." American Journal of Agricultural Economics 60 (1): 65­73. World Development Indicators Database. World Bank. http://go.worldbank.org/B53SONGPA0 (accessed May 2008). 8 VIETNAM Prema-Chandra Athukorala, Pham Lan Huong, and Vo Tri Thanh Since the late 1980s, Vietnam has been making remarkable progress in its transi- tion from a closed command economy toward an open market economy and integration in the world economy. The initial progress was slow and hesitant fol- lowing the announcement of the doi moi (renovation) policy in 1986, but more significant reforms were undertaken in the first half of the 1990s. The process lost momentum in 1996­98, perhaps reflecting complacency after the success of the first reforms, but also the economic uncertainty created by the 1997­98 Asian financial crisis. Since about 1999, there has been a renewed emphasis on complet- ing the unfinished reform agenda. The key reform measures have included widespread agricultural reforms involving a transition from the collective regime to a system in which farmers pos- sess more freedom in making production decisions and in marketing their pro- duce. Quantitative import restrictions have been dismantled in all cases except sugar and petroleum products. Significant tariff reforms have led to notable reductions in the level and dispersion of effective rates of protection. Public sector enterprises have been exposed to greater market discipline. Restrictions have been relaxed on foreign direct investment, particularly in export-oriented projects, and they have been lifted on private sector participation in foreign trade and the establishment of business ventures by private individuals and companies. These initiatives have been accompanied by sweeping macroeconomic policy reforms, including the unification and realignment of the exchange rate, the liberalization of agricultural prices, the relaxation of exchange controls, and a firm commitment to fiscal prudence. 281 282 Distortions to Agricultural Incentives in Asia The purpose of this chapter is to examine the implications of the market- oriented policy reforms in Vietnam for the incentives faced by farmers in the con- text of the changes in the overall structure of the incentives for private sector activ- ities in the economy. We undertake the analysis against the backdrop of a narrative on the evolution of agricultural policy and the key policy trends dating back to the era of the command economy. The empirical analysis of agricultural incentives covers six major products--rice, sugar, pig meat, poultry, rubber, and coffee-- using data from 1986 (the earliest postreform year for which the required data are available) to 2004. The six covered products accounted for more than two-thirds of the total value of agricultural production in Vietnam during the period. Our aim is to contribute to the contemporary policy debate on reforming the structure of incentives for domestic agriculture as an integral part of the endeavor to accelerate the country's economic integration into the world economy. The study has four main parts. The next section presents an overview of the growth and structural changes during the postreform era (since the mid-1980s). It emphasizes the relative importance of the agricultural sector and the trends and compositional shifts in agricultural output and trade. The subsequent section provides an overview of the origins, key elements, and achievements of the effort to meet reform commitments. It highlights the political economy of the related policy making. The penultimate section--the analytical core of the chapter-- examines trends and patterns in the incentives in domestic agriculture by relying on a set of indicators based on the methodology described in Anderson et al. (2008) and appendix A. The final section summarizes the key findings and policy implications. Agriculture in the Vietnamese Economy We first document the extraordinary economic growth performance of this trans- forming economy and then review the structural changes that have accompanied this success. Growth trends During the era of central planning (from the mid-1950s in the north of the coun- try and, in the south, beginning in 1975 following unification), the economy was not subject to forced industrialization to the same degree as the centrally planned economies of China and the Soviet bloc. The prolonged military conflict with the regime in the south and with the United States caused industrial transformation to be limited to the establishment of industries that met the priorities of the war economy. Thus, agriculture--broadly defined to include farming, fisheries, and Vietnam 283 forestry--remained the dominant sector of the economy up to the 1980s. During the period from 1955 to 1985, the share of agriculture in gross domestic product (GDP) fluctuated from around 38 to 52 percent without showing any clear trend (GSO 2001). By the mid-1980s, over 72 percent of the total labor force was engaged in agricultural pursuits (Riedel 1993). The process of the collectivization of agriculture in the northern part of the country had been completed by the early 1960s. The forced replacement of a semisubsistent peasant commodity production system by a regular plan ushered in an era of suppressed growth, if not stagnation, in agriculture. During most of the ensuing three decades, agricultural output in the north was only barely suffi- cient to meet domestic consumption requirements in that part of Vietnam. Attempts to replicate the collectivized system in the south following the adminis- trative unification of the country in 1976 resulted in a severe disruption in agri- cultural production there. Piecemeal reforms implemented in 1979­80 to loosen the structures of central planning had only limited impact in containing output contraction. By the mid-1980s, large areas of the country were experiencing near- famine conditions, and food shortages were resulting in widespread suffering. National food security became a leading preoccupation (White 1985; Riedel and Comer 1997; Pritchett 2003). The response of agriculture to the market-oriented policy reforms launched in the late 1980s was remarkably swift. Between 1988 and 1992, GDP increased by 27 percent; nearly 30 percent of this increase arose directly from agriculture. In addition, rapid agricultural growth contributed to an expansion in rural nonagri- cultural services, input supply, and food processing. During the ensuing years, the growth turned out to be broadly based, and industry and services developed at a pace that was more rapid than the pace of development in agriculture. Nonethe- less, the growth rate in agriculture remained impressive (3.0 to 5.2 percent per year) relative to the precrisis experience in Vietnam and the average performance in other low-income and transition economies. Despite the notable structural changes over the past 15­20 years, agriculture still has a substantial weight in the Vietnamese economy, contributing about 20 percent of GDP (figure 8.1) and absorbing 57 percent of the total labor force in 2005. Slightly more than two- thirds of households in the lowest income quintile were occupied in agriculture in 2004, and almost three-fifths of the incomes among households in that quintile was generated by agricultural activities (compared with less than one-fourth of the incomes in the highest income quintile). The impressive agricultural growth, especially the surge in paddy production, played a key role in winning political support for more reforms by ensuring national food security, a source of much political anxiety in the 1980s. Agricul- tural growth was also at the heart of the success in the rapid reduction in rural 284 Distortions to Agricultural Incentives in Asia Figure 8.1. Agriculture, GDP, and Value Added, Vietnam, 1985­2005 12 50 45 10 40 agriculture 8 35 % agricultural 30 6 and 25 in added, 4 GDP GDP 20 in value ,% 2 15 10 growth 0 5 2 0 198519861987198819891990199119921993199419951996199719981999200020012002200320042005 year agriculture in GDP, % GDP growth agricultural growth Source: Compiled from data in GSO, "Statistical Yearbook," various. poverty. The increase in rural incomes supported the economic transition by reducing the pressure for migration from rural to urban areas despite the widen- ing gap between rural and urban incomes. In contrast to China, internal migra- tion in Vietnam has occurred as much from one rural area to another as from the countryside to the city. This has limited the demand for substantial expenditure on urban development (Minot and Goletti 2000; Van Arkadie and Mallon 2003). The production of major commodities Paddy and rice were the prime movers of agricultural growth during the immedi- ate postreform period. Since the mid-1990s, agricultural production has been diversifying significantly into other food crops (maize, peanuts, and soybeans), cash crops (particularly rubber, coffee, tea, cashews, pepper, and cinnamon), fruits and vegetables, marine and aquaculture products (shrimps, fish, cuttlefish, and crabs), and animal husbandry (pig meat and poultry). In cash crops such as coffee, cashews, and pepper, there has been a shift from negligible production to sufficient production to make the country a major player in the relevant world markets. The initial expansion in the production of some cash crops (especially rubber) reflected returns to government investment in state farms in the 1980s, Vietnam 285 Table 8.1. Structure of the Agricultural Economy, Vietnam, 1986­2004 (percent) Indicator, product 1986­89 1990­94 1995­99 2000­04 Agriculture in GDP 42 34 26 23 Composition of agricultural output -- 100.0 100.0 100.0 Rice -- 46.6 45.4 35.7 Rubber -- 1.6 2.0 2.3 Coffee -- 2.1 5.2 3.8 Sugar -- 3.5 4.4 3.4 Pig meat -- 6.4 7.3 9.9 Other -- 39.7 35.8 45.0 Source: Compiled from data in GSO, "Statistical Yearbook," various. Note: The estimates have been calculated by applying value added shares, at current prices, from the input-output tables in 2000 to gross output data. -- no data are available. but the growth in agricultural production during the postreform era has arisen predominantly from smallholder production in the private sector. Rice, the staple food of the country, accounts for three-quarters of the calorie intake among the population and is the dominant agriculture product. In 2004, paddy accounted for 57 percent of total cultivated area and 36 percent of total agricultural output (tables 8.1 and 8.2 and figure 8.2). The Red River Delta and the Mekong River Delta account for around one-third and more than half of the total national production of paddy, respectively, although paddy, the primary food crop, is also grown widely in all other parts of the country. Paddy production increased steadily from 19.2 million tons in 1990 to 36.1 million tons in 2005. The impressive annual compound growth rate of 4.2 percent was generated largely because of an improvement in yield per hectare, while the area under paddy remained virtually unchanged (Athukorala, Huong, and Thanh 2007). Paddy yields rose from 2.8 tons to 4.9 tons per hectare between 1986 and 2005. Coffee, rubber, and sugarcane are the three most important cash crops. In 2004, coffee accounted for 3.8 percent of agricultural output, while sugarcane and rubber accounted for 3.4 and 2.3 percent, respectively. Coffee and rubber produc- tion is largely aimed at export markets; the sugar industry generally meets only domestic demand. Coffee and sugar are predominantly smallholder crops. Rubber is mainly produced on farms owned by state-owned enterprises at the provincial level or by the General Rubber Corporation, a state-owned enterprise at the national level. The area of cultivation under rubber and coffee and the yield and overall output levels of these two products have recorded impressive growth since the early 1990s (Athukorala, Huong, and Thanh 2007). Sugar production jumped 286 Distortions to Agricultural Incentives in Asia Table 8.2. Share of Planted Area by Crop, Vietnam, 1990­2004 (percent) Crop 1990 1995 2000 2004 Paddy 66.8 64.5 60.6 56.6 Maize 4.8 5.3 5.8 7.5 Sugar 1.4 2.1 2.4 2.2 Groundnuts 2.2 2.5 1.9 2.0 Soybeans 1.2 1.2 1.0 1.4 Tea 0.7 0.6 0.7 0.9 Coffee 1.3 1.8 4.4 3.8 Rubber 2.5 2.7 3.3 3.4 Pepper 0.1 0.1 0.2 0.4 Coconuts 2.3 1.6 1.3 1.0 Total 100.0 100.0 100.0 100.0 Total area, hectares, 1,000s 9,040 10,497 12,644 13,150 Source: Compiled from data in GSO, "Statistical Handbook," various. Figure 8.2. Commodity Shares in Agricultural Production, Vietnam, 1991­2002 100 90 at 80 % 70 60 prices, production 50 of 40 share 30 undistorted 20 value 10 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 year noncovered rubber coffee sugar chicken pig meat rice Source: Compiled from data in GSO, "Statistical Yearbook," various. Note: The figure is based on estimates relying on data in the input-output tables in 2000. Vietnam 287 in 1995 following the launch of the One Million Tons of Sugar Program and con- tinued to increase up to 1999. However, a mild downward trend characterized by significant annual fluctuations has been evident since then. Sugar yields rose from 396 to 553 quintals per hectare in 1986­2005. The area under sugarcane cultiva- tion has declined in recent years because of a shift by farmers to other crops, mostly subsidiary food crops. Despite its relatively poor performance (or because of it), sugarcane production remains the most well assisted agricultural activity (see below). In 1995, the government launched the One Million Tons of Sugar Program, which was aimed at achieving sugar self-sufficiency by 2000 and creat- ing employment in the rural economy (Nguyen et al. 2006). The other cash crops that have shown impressive growth during the postre- form era include cashews, groundnuts, tea, and pepper. Vietnam is the world's largest producer of pepper, the third largest producer of cashew nuts, the fifth largest producer of tea, and the tenth largest producer of groundnuts. However, the combined share of these products in the total agricultural GDP of the country remains small, at less than 3 percent. Livestock production has expanded rapidly since the early 1990s. It accounted for about 14 percent of agricultural value added in 2000 (IAPP 2001). Pig meat is the most important livestock product (60 percent), followed by poultry (15 per- cent) and beef (8 percent). The share of pig meat in agricultural value added rose from 6 percent in 1990­94 to 10 percent in 2000­04 (table 8.1). Currently, over 90 percent of total pig meat production is consumed domestically, although exports (mainly to China) have been increasing rapidly in recent years. Agricultural exports Primary products accounted for nearly half of non-oil merchandise exports in the mid-1980s. The share rose in the early years of the postreform period when the first positive response to the reforms was registered in agricultural products, espe- cially rice. In 1998, Vietnam became self-sufficient in rice and then a net exporter. In 2005, rice exports peaked at 5.3 million tons, making the country the third largest rice exporter in the world (after the United States and Thailand). The com- position of agricultural exports has become increasingly diversified; pepper, cashews, rubber, coffee, and fish products have shown impressive growth. Since the late 1990s, manufacturing exports have been growing more rapidly, resulting in a notable shift in the composition of exports away from primary products. However, agricultural products still accounted for over one-quarter of total non- oil merchandise exports in 2004 (table 8.3).1 Until about the mid-1990s, rapid volume expansion and favorable price trends were contributing to the growth in agricultural export earnings (figure 8.3). Prices 288 Distortions to Agricultural Incentives in Asia Table 8.3. Composition of Agricultural Exports by Value, Vietnam, 1990­2004 (percent) Indicator, product 1990 1995 2000 2004 Agricultural exports in total non-oil exports 80 46 25 22 Composition of agricultural exports 100.0 100.0 100.0 100.0 Groundnuts -- 3.7 2.3 0.9 Rubber 4.7 12.0 9.4 20.5 Coffee 7.3 37.4 28.4 22.0 Tea 0.6 0.8 4.0 3.3 Rice 80.2 40.7 37.8 32.7 Cashews 3.8 9.8 9.5 15.0 Black pepper 3.5 4.5 8.3 5.2 Cinnamon -- -- 0.3 0.3 Source: Compiled from data in GSO, "Statistical Yearbook," various. Note: -- no data are available. Figure 8.3. Volume, Value, and Price Indexes, Agricultural Exports, Vietnam, 1990­2004 900 800 700 100 600 = 500 1990 400 300 index, 200 100 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 year value price volume Source: Athukorala, Huong, and Thanh 2007. continued to decline thereafter; the rate of decline has intensified in recent years. Rapid volume expansion compensated for the price decline until about 1998, gen- erating mild, but positive growth in export earnings. However, the rate of growth in agricultural export earnings has steadily slowed over the past five years or so; this mostly reflects declining prices. Vietnam 289 The Reform Process: From Plan to Market Under the collective system of agriculture instituted in the north in the early 1960s, cooperatives were the key link between agricultural households and the national economic plan. As the prime institution in the effort to replace the mar- ket by the plan, the agricultural cooperatives were responsible for organizing the deployment of the agricultural labor force, undertaking production according to plans approved by the central authorities, selling surplus production to the state at state-controlled prices, and implementing obligatory procurement quotas intro- duced from time to time for sales to the state of selected essential commodities (White 1981). After the defeat of the government in the south in 1975 and the formal admin- istrative reunification of the economy in 1976, replacing the market by the plan in the south presented a formidable challenge. Immediately after reunification, the attempt to bring southern agriculture into the collective system was fairly cau- tious. However, the rapid growth in private trade, combined with concerns about the political resistance to socialist transformation among southern farmers and the southern business community, which was dominated by ethnic Chinese, led to an attempt to accelerate the process. The Second Party Plenum, in July 1977, set ambitious targets to speed the collectivization of individual farm households. The resistance among farmers to the introduction of the collective system, coupled with uncertainty about the future direction of reform, resulted in declines in agri- cultural output. The process of collectivization in the south was slowed. Agree- ments were then reached on the need to decentralize decision making and provide better incentives for increasing production though private household initiatives (Duiker 1989; Fforde and de Vylder 1996; Naughton 1996). Reforms in 1979­80 introduced production contracts under which coopera- tives subcontracted land to households and allowed households greater latitude in decision making on production issues. Under this new system, which was similar to the household responsibility system in China, households were allowed to keep or to sell on the free market any surplus above the amount stipulated for delivery to cooperatives under the contracts. In effect, the cooperatives were limited to the subsidiary role of allocating land, supplying inputs, and providing technical assis- tance (Woodside 1989). These reforms had an immediate and dramatic effect: in 1994 prices, total agricultural production rose from D 37 trillion in 1979 to D 46 trillion in 1982 (GSO 2001). However, rather than responding to improved production by ampli- fying reforms, as the Chinese government did, the Vietnamese government backpedaled for most of the rest of the decade. The emergence in the early 1980s of severe macroeconomic imbalances, which were reflected in high and rising inflation, undermined the reform movement. The macroeconomic problems were 290 Distortions to Agricultural Incentives in Asia interpreted as a symptom of the failure of the reforms, and they also created dis- satisfaction because they led to a reduction in real wages in the civil service. Thus, the influence of hard-liners within the Communist Party of Vietnam had gained strength by the mid-1980s, and this added to the pressure to enforce the collec- tivization of agriculture in the south (Riedel and Comer 1997). By the mid-1980s, the economy was stagnating because of hyperinflation and a chronically poor balance of payments situation. Furthermore, it was clear by 1988 that Soviet aid would soon decline. In the face of these problems, a more con- certed push toward reform was announced at the Sixth Congress of the Commu- nist Party of Vietnam in December 1986 (CPV 1994). The implementation of this program of reform, referred to as doi moi, did not, however, gain much momen- tum until the collapse of the Soviet Union, which put an end to Soviet aid. Massive contraction in agricultural output in 1988, which brought near-famine conditions in many parts of the country, also played a role in reducing the resistance to reform. The focus of early reform was mainly on unshackling agriculture. Initially, the reform process largely ignored the private sector except in agriculture, and the process of establishing the institutions needed to support private sector activity outside agriculture was undertaken only in the early 1990s after emerging signs of unrest. Unshackling agriculture The transition to a more decentralized, market-oriented system of agricultural production was launched through the adoption of a decree by the Communist Party of Vietnam in 1988. The decree recognized the peasant household rather than the cooperative as the basic unit in the agrarian structure. It recognized the right of households to the conditional use of private land for a period of 10­15 years; to own draft animals, farm tools, and other equipment; to barter output for inputs; and to retain the income earned through production once they had paid a modest tax. However, cooperatives continued to have ultimate control over land and water resources; and the sale of output, at prices set by the government, remained restricted to sales within the same district. Additional measures introduced in 1989 reduced the direct involvement of the government in the allocation of inputs. In July 1993, the duration of tenure rights over agricultural land was extended to 20 years, and farmers were permitted to sell, lease, exchange, mort- gage, and bequeath land. Cooperatives were still meant to provide a focus for various rural activities sponsored by the government, but, in the majority of com- munes, they were reduced to only a minor role. They were to act as local tax collectors, the holders of residual property rights, and a component of the formal state structure (Riedel 1993; Sachs and Woo 1994; Riedel and Comer 1997). Vietnam 291 Land tenure reforms were accompanied by sweeping domestic market price reforms. In 1987­88, the rationing system was abolished for many commodities, and the official prices of nonessential goods were raised to a level close to free- market prices. Administered prices on most consumer goods and a large number of agricultural and industrial inputs were abolished. In June 1990, the procure- ment of farm products by the government, usually at prices below free-market prices, formally came to an end, allowing farmers to sell their produce at mar- ket prices. By 1990, commodity prices were largely determined by domestic mar- ket conditions, and direct subsidies had been eliminated. The sellers market was replaced because of the shift toward market-clearing prices. The weakening of the state trading system at the local level permitted private traders to develop local markets, while many state trading enterprises became more responsive to market opportunities. Starting in 1989, the international trade in agricultural products was also liberalized in phases, and this eventually allowed private sector participation. Trade policy reforms Foreign trade and investment regimes were liberalized in stages in conjunction with the domestic market price reforms. The law on import and export duties introduced in January 1988 marked the launch of the new trade tax system. The original import tariff schedule was replaced in 1992 by a detailed, consolidated schedule based on the Harmonized Commodity Description and Coding System of tariff nomenclature. During the rest of the 1990s, the tariff structure was fine- tuned to mirror a policy trend toward selective protection in the case of consumer goods (cosmetics and several categories of food products), activities upstream of the textile and garment industry (silk, cotton, and some fiber production), and some intermediate goods (metal products, cement, and glass). In the early and mid-2000s, following Vietnam's accession to the free trade area of the Association of Southeast Asian Nations in 1995 and in preparation for the country's accession to the World Trade Organization (WTO), the government took steps to reorgan- ize and rationalize the tariff structure (Thanh 2006). After more than 15 years of reform, tariffs are now the major instruments used in regulating import trade. The average (import-weighted) import duty rate declined from 22 percent in 1999 to 13.6 percent in 2004 (figure 8.4). The maxi- mum tariff rate (at the six-digit level of the Harmonized Commodity Description and Coding System) declined from 200 percent in 1997 to 120 percent in 2001 and then 113 percent in 2004. By October 2005, of all tariff lines, less than 1 percent (accounting for around 4 percent of total import value) were subject to tariff rates above 50 percent. The tariff on about one-third of the tariff lines was zero. Despite 292 Distortions to Agricultural Incentives in Asia Figure 8.4. Weighted Average Import Duties, Vietnam, 1990­2004 25 20 % rate, 15 tariff 10 import 5 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 year Source: Nicita and Olarreaga 2006; CIE 1998. notable efforts to rationalize the tariff structure, tariffs are still relatively high and nonuniform according to regional standards (Athukorala 2006).2 The tariff rates are generally higher for manufacturing products than for agri- cultural products and other primary products. By mid-2003, the weighted average duty rate on manufacturing imports was 29 percent, while the corresponding rates were 11 percent on agricultural products and 3.6 percent on mineral prod- ucts. Within manufacturing, the tariff rates are particularly high for processed foods and some consumer goods, notably garments, footwear, ceramic products, and leather goods (Athukorala 2006; Athukorala, Huong, and Thanh 2007). By 2004, only two products, sugar and petroleum, remained subject to quanti- tative restrictions (import licensing). As part of its trade reform commitments under the WTO accession agreement, the government offered to replace import licensing on sugar by a WTO-consistent tariff trade quota system. Imports of two products--poultry eggs and raw tobacco--were already subject to tariff rate quotas. The list of prohibited imports included military equipment, toxic chemi- cals, antiquities, narcotics, firecrackers, toxic toys, used consumer goods, and right-hand-drive automobiles. In addition, a considerable number of import items (such as pharmaceuticals, some chemicals, some food items, fertilizer, and recording and broadcasting equipment) required approval from the relevant min- istries. In 2000, in value terms, around 10 percent of imports were subject to this Vietnam 293 form of regulation. As in many other countries, these regulations are generally maintained in Vietnam for heath and security reasons, and they do not appear to distort trade patterns substantially. Among the early market-oriented reforms, the government introduced duties on a number of export items. The duties were justified at the time by the need to raise revenue, protect the environment, conserve natural resources, and retain inputs for domestic production. Most of the duties were subsequently eliminated. By 1998, only a few export products--iron ore, crude oil, scrap metal, and raw cashews--were subject to duties. Export prohibitions now apply only to environ- mentally sensitive items such as agroforestry products, round wood and saw wood cut from domestic natural forests, firewood and charcoal obtained from natural- growth domestic forestry wood, and rare wild animals. When the country began exporting rice in 1989, the exports were subject to licensing to ensure adequate domestic supplies and reduce the price volatility in the domestic market. Export quotas were issued only to a few state-owned enter- prises. During the next years, the number of enterprises affected ranged from 15 to 40. The intense lobbying by these enterprises to share in the export quotas dur- ing the early years of the rice exports suggests that the quotas were binding and ensured that the domestic price was below the relevant border price. However, starting in about 1998, the quotas became ineffective because of the rapid expan- sion in rice production and, thus, in the marketable surplus. In April 2001, the rice export quota allocation mechanism, together with the import quota system for fertilizer, was abolished by a prime ministerial decision. Enterprises were thence- forth permitted to export rice, provided they had obtained general business licenses to trade in rice or other agricultural products. For rice exports to coun- tries with which the government has bilateral trade agreements, the Ministry of Trade assigns export rights to selected enterprises in consultation with the Vietnam Food Association. However, this trade is too small to have a significant impact on domestic rice markets. Other reforms The reforms in agriculture and foreign trade were accompanied by significant macroeconomic policy reforms (Dollar 1992; Dollar and Ljunggren 1997). To fight inflation, interest rates were raised to high levels. The government also tried to curb deficit financing, which required a large fiscal adjustment that led, for example, to the release of 500,000 soldiers from the military and sharp cuts in the subsidies to state-owned enterprises. These policy measures, combined with revenue windfalls from petroleum operations that had come on line, brought the budget deficit down from 11.4 to below 4 percent of GDP in 1989­92. Fiscal adjustment and monetary 294 Distortions to Agricultural Incentives in Asia Figure 8.5. Index, Real Exchange Rate, Vietnam, 1988­2005 300 100 250 = 200 2000 rate, 150 100 exchange 50 real 0 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 year Source: Athukorala, Huong, and Thanh 2007. restraint were successful in reducing the inflation rate from over 160 percent a year in 1988 to less than 10 percent a year by the mid-1990s. Exchange rates were unified, and the new rate was sharply devalued in 1989. The real exchange rate devaluation amounted to 72.5 percent according to the International Monetary Fund (Dollar 1992). Since then, the Vietnamese dong (D) has been on a managed floating exchange rate regime whereby the State Bank of Vietnam (the central bank) determines the unified rate according to foreign exchange market trading. In 1990­98, the gap between the official exchange rate and the exchange rate in the interbank market varied from 5 to 10 percent. Since then, the approach of the state bank to exchange rate management has become more flexible, and the gap between the two rates has been reduced to around 0.1 percent on any given business day. The black market premium on the U.S. dol- lar, which stood at over 50 percent in 1988­95, had fallen to less than 5 percent by 2004 (Athukorala, Huong, and Thanh 2007). Because of successful macroeco- nomic stabilization and exchange rate management, the real exchange rate has remained stable since the mid-1990s (figure 8.5). Trends and Patterns in Agricultural Incentives In this section, we provide an analysis of the changing scope of and patterns in the direct and indirect distortions in the incentives faced by domestic agriculture in Vietnam 295 Vietnam. We rely on the methodology developed by Anderson et al. (2008) and described in appendix A. The main focus of our use of the methodology is government-imposed distortions that create a gap between domestic prices and the prices that would emerge under free market conditions. We estimate the effects of direct agricultural policy measures, but, for comparative evaluation and because it is not possible to understand the characteristics of agricultural develop- ment through a sectoral view alone, we also estimate the distortions in nonagri- cultural tradables. Specifically, we compute nominal rates of assistance (NRAs) for farm producers of the six products we cover. We also calculate a relative rate of assistance (RRA) based on the NRAs for nonagricultural tradables and the NRAs for agricultural tradables. We assume that the NRAs for noncovered agricultural exportable products are the same as the average NRAs for covered exportables, and we assume that the average NRA for noncovered import-competing products is 10 percent of the NRAs for sugar. We also assume that the average NRA for nontradables is zero and that the share of nontradables in noncovered farm pro- duction is 68 percent, while the exportable share is 25 percent. We have estimated the NRAs for nonagricultural tradables by assuming that the implicit duty rate on nonagricultural importables (the total tariff revenue generated from nonagricultural imports, divided by the value of nonagricultural imports) is the rate of distortion in this component of nonagricultural tradables and that nonagricultural exportables (the share of which is three-quarters as large as the share of import-competing products) are not subject to any export taxes or subsidies. One should bear in mind two important caveats that arise from the paucity of data. First, we have ignored potential differences between border (reference) prices and domestic prices that arise from quality differences. This might have infused an underestimation bias into our calculations. Second, we have assumed that there is a complete pass-through of the changes in producer (wholesale) prices to farmgate prices. This may result in an upward bias in the estimates. How- ever, these limitations are important only in comparisons of the levels of distor- tion in the incentives among products or across countries at a given time. They are unlikely to distort inferences based on intertemporal comparisons (changes in incentives over time) because the magnitude of the bias is less liable to vary over time. Given the nature of the estimation method, our RRA estimates also do not fully capture the indirect distortions that arise in agricultural incentives from changes in the tariffs on tradable inputs. The tariff structure in Vietnam is cascad- ing. This may be an important source of downward bias in our RRA estimates (Athukorala 2006). We have also not captured the distortions caused by nontariff import restrictions. 296 Distortions to Agricultural Incentives in Asia Table 8.4. NRAs for Covered Agricultural Products, Vietnam, 1986­2004 (percent) Indicator, product 1986­89 1990­94 1995­99 2000­04 Exportablesa 13.2 27.2 2.1 16.9 Rice 2.8 26.6 0.4 22.9 Rubber -- 21.2 18.6 16.8 Coffee 49.4 21.1 7.1 12.0 Pig meat 41.8 37.5 6.1 8.9 Poultry 3.1 3.6 3.7 1.6 Import-competing productsa -- 49.6 112.9 160.2 Sugar -- 49.6 112.9 160.2 All covered productsa 13.2 27.2 0.2 20.6 Dispersion, covered productsb 28.8 46.1 157.7 221.3 Coverage, at undistorted prices 70 67 76 63 Source: Athukorala, Huong, and Thanh 2007. Note: -- no data are available. a. Weighted averages; the weights are based on the unassisted value of production. b. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of covered products. Our estimated distortion rates are summarized in five-year averages in tables 8.4 and 8.5 and are plotted in figure 8.6. Detailed annual estimates, the esti- mated NRA series for each of the covered products, and the related domestic and border price series are reported by Athukorala, Huong, and Thanh (2007). In the late 1980s and the first half of the 1990s, the policy regime was characterized by a significant bias against agriculture. The RRA averaged 20 percent in 1990­94. The direct negative assistance to agriculture (as measured by the NRAs for agricul- ture) underpinned this high degree of distortion in agricultural incentives. The main factors that kept domestic prices artificially suppressed relative to border prices were the continued dominance of state-owned enterprises in the trading and processing of agricultural commodities; a stringent export licensing system in the rice trade; other trade restrictions; administered prices, which were usually main- tained below border prices; and, perhaps, a lack of experience among the emerging private traders who were operating in the newly competitive trading environment. Until the mid-1990s, four exportable agricultural products--rice, coffee, pig meat, and poultry--faced significant negative assistance (table 8.4). Rubber is unique among the five exportables because it has enjoyed positive assistance over the last two decades. This reflects production subsidies for the state-owned plantation companies that account for the bulk of rubber production. In the Vietnam 297 Table 8.5. NRAs in Agriculture Relative to Nonagricultural Industries, Vietnam, 1986­2004 (percent) Indicator 1986­89 1990­94 1995­99 2000­04 Covered productsa 13.2 27.2 0.2 20.6 Noncovered products 14.5 25.0 0.3 22.3 All agricultural productsa 14.0 26.5 0.1 21.2 Non-product-specific assistance 0.0 0.0 0.0 0.0 Total agricultural NRAb 14.0 26.5 0.1 21.2 Trade bias indexc 0.19 0.17 0.01 0.00 NRA, agricultural tradables 16.1 26.4 0.0 20.7 NRA, nonagricultural tradables 4.3 11.2 1.5 20.8 RRAd 19.4 17.4 1.3 0.0 Source: Athukorala, Huong, and Thanh 2007. a. Including product-specific input subsidies. b. Including product-specific input subsidies and non-product-specific assistance. Represents total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. c. The trade bias index is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. d. The RRA is defined as 100*[(100 NRAagt ) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. mid-1990s, after the liberalization of the export trade and the removal of direct price interventions and some export duties, the incentives for rice, pig meat, and poultry were significantly improved. Even though coffee production remained disprotected (taxed), the overall NRA index for exportable agriculture rose steadily in the ensuing years from around 27 percent in 1990­94 to around 17 percent in 2000­04. Sugar, the only import-competing product covered in our study, occupies a singular position because of the trend in assistance. Sugarcane producers enjoyed NRAs that were exceptionally high relative to the NRAs among producers of other products, and the measured level of assistance increased steadily. In 2000­04, the NRA for sugarcane was 160 percent, while the weighted average was 20 percent for all covered products. This points to the stringency of the licensing regime govern- ing sugar imports. The government's policy of protecting sugar has been system- atically analyzed in several recent studies (for instance, see CIE 2001; Nguyen et al. 2006). These studies conclude that the government's sugar industry development strategy, which is enshrined in the One Million Tons of Sugar Program launched in 1995, has been a dismal failure and that a competitive, economically viable 298 Distortions to Agricultural Incentives in Asia Figure 8.6. NRAs for Agricultural and Nonagricultural Tradables and the RRA, Vietnam, 1986­2004 40 30 20 10 % 0 NRA, 10 20 30 40 50 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 year NRA, nonagricultural tradables NRA, agricultural tradables RRA Source: Athukorala, Huong, and Thanh 2007. Note: For the definition of the RRA, see table 8.5, note d. sugar industry cannot be developed through isolation from world market conditions. The substantial protection provided to the sugar industry is a major constraint on the diversification of scarce land resources to more dynamic, export-oriented crops. Moreover, the high domestic sugar prices not only tax domestic consumers, but also hamper the competitiveness of the domestic con- fectionary, food, and beverage industries. Improvement in the NRAs for exportable agriculture, coupled with the mainte- nance of the high NRAs enjoyed by sugarcane producers, have generated an improvement in the RRAs for agriculture over the past 10 years. Moreover, this improvement in the relative incentives in agriculture has taken place against the backdrop of a steady rise in the NRAs for nonagricultural sectors over time (table 8.5 and figure 8.6). Clearly, the elimination of various direct price interventions in domestic trade, the removal of export duties on almost all agricultural products, the liberalization in import trade, and the exchange rate reforms have been instrumental in redressing the antiagricultural bias in the incentive structure within the economy. Vietnam 299 A comparison of the weighted average NRAs for the covered exportables (rice, rubber, coffee, pig meat, and poultry) with the corresponding NRAs for sugar (an import-competing product) points to a persistent bias in agricultural incen- tives in favor of import-competing production in agriculture (table 8.4). How- ever, this comparison needs to be qualified: the NRAs for import-competing products refer only to the protection for sugar, which is an outlier relative to other import-competing agricultural products in Vietnam that are not covered in this study. Concluding Remarks Over the past two decades, Vietnam has made significant progress in market- oriented reforms. The foreign trade regime has become more liberalized; there has been a palpable transition from quantitative restraints to tariffs as the main instrument in regulating imports. Export taxes on all significant products have been eliminated. In domestic trade, the dominance of state-owned enterprises in most areas has ended, and price controls and restrictions on production and the movement of goods no longer exist. The reform process is far from complete, however. The structure of trade pro- tection is still out of step with the level and the dispersion of nominal and effec- tive protection rates among Vietnam's major trading partners in the region. Stringent quantitative import restrictions on sugar and high import duties on agricultural products in which the country has a clear comparative advantage, particularly rice, coffee, and tea, are major anomalies. The tariff rates are exces- sive on crucial agricultural inputs and other intermediate goods that are pro- duced locally by state-owned enterprises. Had information on these rates been available to us in our analysis, our NRA estimates in agriculture would have been lower. The licensing scheme for rice exports, although seemingly nonbinding for some time now, remains an important source of uncertainty among private sector traders. The attempt to unshackle domestic agriculture accounted for the first of the market-oriented reforms, and the reforms in this area have been more wide- ranging than the reforms in other areas. The predominance of agriculture in the prereform economy--its importance in determining the fortunes of the economy and ensuring the livelihoods of a majority of the population--made the sweeping agricultural reforms politically palatable. Because the vast potential of agriculture was untapped under the command economy, the response of agriculture to the reforms was swift. The impressive outcome played a pivotal role in sustaining the momentum of the reforms and supporting market-oriented programs. 300 Distortions to Agricultural Incentives in Asia Our empirical analysis of trends and patterns in the incentives in agriculture has yielded several important findings. In the first half of the 1990s, the policy regime was characterized by a significant bias against agriculture. This bias dissi- pated because of the gradual removal of the privileges enjoyed by state-owned enterprises in procuring, processing, and trading agricultural products; the open- ing of domestic and foreign trade to the private sector; and the emergence of the nascent private sector in a competitive trade environment. By 2000­04, the NRAs in agriculture and the degree of antiagricultural bias embodied in the overall poli- cies affecting tradables represented evidence that the bias had been reversed. The improvement in the relative incentives for agriculture was generated predomi- nantly, if not solely, by direct reforms in agriculture. This is demonstrated by the small size of the increase in the NRAs for nonagricultural tradables. In this con- text, the implementation of the tariff reform commitments undertaken during the accession to the WTO may play a vital role in consolidating the reform effort and reducing the remaining sectoral policy bias. According to our commodity-level estimates, the excessive assistance provided to sugar producers, mainly though stringent quantitative restrictions on sugar imports, is the major irregularity in the incentive structure. The NRAs have con- tinued to be high in recent years despite a steady decline in border prices; this reflects the severity of the existing quantitative restrictions. The significant protec- tion provided to this industry is a major constraint on the diversification of agri- culture into dynamic export-oriented crops. High domestic sugar prices tax domestic consumers, but also hamper the competitiveness of the domestic con- fectionary, food, and beverage industries. Redressing this anomaly in the incentive structure remains a formidable challenge because sugarcane has long been a choice crop in the government's rural development and agricultural diversifica- tion programs. Unfortunately, the government has missed the opportunity to make use of the WTO accession process to face down the political resistance to reform. Instead, it has chosen the soft option of replacing the sugar import licens- ing scheme by WTO-consistent tariff rate quotas during 2008. Notes 1. Export shares are estimated in gross terms, that is, on the basis of published trade data without adjusting for import content. This tends to understate the balance of payments implications of agri- cultural exports, particularly because most of the newly emerging manufactured exports greatly depend on imported inputs. 2. By mid-2003, the average unweighted tariff rate, 16.7 percent, was aa little lower than the average rates in China (17.5 percent) and Thailand (18.5 percent), but much higher than the rates in Indonesia (8.4 percent), Malaysia (10.2 percent), and the Philippines (7.6 percent). The degree of dispersion in the tariff rates--measured through the coefficient of variation--is much higher in Vietnam than in China, the Philippines, and Thailand, but lower in Vietnam than in Indonesia and Malaysia. Vietnam 301 References Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Athukorala, P.-C. 2006. "Trade Policy Reforms and the Structure of Protection in Vietnam." World Economy 29 (2): 161­87. Athukorala, P.-C., P. L. Huong, and V. T. Thanh. 2007. "Distortions to Agricultural Incentives in Vietnam." Agricultural Distortions Working Paper 26, World Bank, Washington, DC. CIE (Center for International Economics). 1998. Vietnam's Trade Policies 1998. Canberra: CIE. ------. 2001. Vietnam Sugar Program: Where Next? Canberra: CIE. CPV (Communist Party of Vietnam). 1994. "Political Report of the Central Committee to the Midterm Party Conference." Report, CPV, Hanoi. Cited in Dollar and Ljunggren 1997, 464. Dollar, D. 1992."Vietnam: Successes and Failures of Macroeconomic Stabilization." In The Challenge of Reform in Indochina, ed. B. Ljunggren, 207­31. Cambridge, MA: Harvard University Press. Dollar, D., and B. Ljunggren. 1997. "Vietnam." In Going Global: Transition from Plan to Market in the World Economy, ed. P. Desai, 439­71. Cambridge, MA: MIT Press. Duiker, W. J. 1989. Vietnam since the Fall of Saigon. 3rd ed. Athens, OH: Ohio University Press. Fforde, A., and S. de Vylder. 1996. From Plan to Market: The Economic Transition in Vietnam. Boulder, CO: Westview Press. GSO (General Statistics Office). 2000. Statistical Data of Vietnam Agriculture, Forestry, and Fishery, 1975­2000. Hanoi: Statistical Publishing House. ------. 2001. Vietnamese Economy during the Years of Renovation in Aggregated Economic Indicators of the System of National Accounts. Hanoi: Statistical Publishing House. ------. various years. Statistical Yearbook of Vietnam. Hanoi: Statistical Publishing House. ------. various years. Statistical Handbook of Vietnam. Hanoi: Statistical Publishing House. IAPP (Institute for Agricultural Planning and Projecting). 2001. "Report on Strategy for Developing the Livestock Sector in Vietnam until 2010." Unpublished study, IAPP, Hanoi. Cited in Nguyen and Grote 2004, 13. Minot, N., and F. Goletti. 2000."Rice Market Liberalization and Poverty in Viet Nam." Research Report 114, International Food Policy Research Institute, Washington, DC. Naughton, B. 1996. "Distinctive Features of Economic Reform in China and Vietnam." In Reforming Asian Socialism: The Growth of Market Institutions, ed. J. McMillan and B. Naughton, 273­96. Ann Arbor, MI: University of Michigan Press. Nguyen, Hoa, and Ulrike Grote. 2004. "Agricultural Policies in Vietnam: Producer Support Estimates, 1986­2002." MTID Discussion Paper 79, Markets, Trade, and Institutions Division, International Food Policy Research Institute, Washington, DC. Nguyen M. T., D. Harris, T. Chong Thang, and N. Q. Nguyen. 2006."Building a Roadmap for the Inter- national Integration of Vietnam's Sugar Industry." Unpublished study, Ministry of Agriculture and Rural Development and Australian Agency for International Development, Hanoi. Nicita, A., and M. Olarreaga. 2006. "Trade, Production, and Protection, 1976­2004." World Bank Economic Review 21 (1): 165­71. Pritchett, L. 2003. "A Toy Collection, a Socialist Star, and a Democratic Dud?" In In Search of Prosperity: Analytic Narratives on Economic Growth, ed. D. Rodrik, 123­51. Princeton, NJ: Princeton University Press. Riedel, J. 1993. "Vietnam: On the Trails of the Tigers." World Economy 16 (4): 401­22. Riedel, J., and B. Comer. 1997. "Transition to a Market Economy in Vietnam." In Economies in Transi- tion: Comparing Asia and Eastern Europe, ed. W. T. Woo, S. Parker, and J. D. Sachs, 189­213. Cambridge, MA: MIT Press. Sachs, J. D., and W. T. Woo. 1994. "Experiences in the Transition to a Market Economy." Journal of Comparative Economics 18 (3): 271­75. Thanh, V. T. 2006."Vietnam's Trade Liberalization and International Economic Integration: Evolution, Problems and Challenges." ASEAN Economic Bulletin 22 (1): 75­91. 302 Distortions to Agricultural Incentives in Asia Van Arkadie, B., and R. Mallon. 2003. Vietnam: A Transition Tiger? Canberra: Asia Pacific Press. White, C. 1981. "Agrarian Reform and National Liberation in the Vietnamese Revolution, 1920­1957." PhD dissertation, Cornell University, Ithaca, NY. ------. 1985. "Agricultural Planning, Pricing Policy and Co-operatives in Vietnam." World Develop- ment 13 (1): 97­114. Woodside, A. 1989. "Peasants and the State in the Aftermath of the Vietnamese Revolution." Journal of Peasant Studies 16 (4): 283­97. Part iv II SOUTH ASIA 9 BANGLADESH Nazneen Ahmed, Zaid Bakht, Paul A. Dorosh, and Quazi Shahabuddin The government of Bangladesh has substantially liberalized its trade and agricul- tural pricing policies since the country's independence in 1971. The government had removed most of the distortions in agricultural incentives by the mid-1990s. Although it has raised the level of trade protection for some agricultural and industrial products sharply since 1998, the total distortions in agriculture remain small. In particular, relative to India and Pakistan, domestic and international trade policies for the major staples--rice and wheat--are substantially more lib- eral in Bangladesh. In the early 1970s, the government pursued highly restrictive trade and exchange rate policies through import regulations, high import tariffs, export taxes, pervasive quantitative restrictions, and an overvalued exchange rate. These polices were similar to the policies implemented in the 1960s when Bangladesh was part of Pakistan. The policy regime in the 1970s was especially restrictive in the agricultural sector. The government had a monopoly on the imports of most agricultural commodities and placed stiff export restrictions on raw jute, the major agricultural export. Because of these distortions, agricultural price incen- tives were substantially reduced during the period (Rahman 1994). Under pressure from donors and disenchanted by the slow economic growth and continued balance of payments problems despite the policies, the government initiated reforms in the early 1980s. Broad liberalization was not undertaken until the early 1990s, however. The liberalization was accompanied by agricultural trade and pricing reforms. By the mid-1990s, distortions in the output prices for rice and wheat had been virtually eliminated, and the total distortions were mini- mal. The government raised the import tariffs on rice in response to subsidized 305 306 Distortions to Agricultural Incentives in Asia exports by India in 2001, but domestic rates of assistance calculated relative to international market prices indicate that there have been only small overall agri- cultural price distortions since around 2000. In the next section, we describe growth and structural change in the economy of Bangladesh. We focus on the agricultural sector. In the subsequent section, we offer an overview of the evolution of agricultural policies since independence. We then report our time series estimates of nominal rates of assistance (NRAs) for selected agricultural products. In the penultimate section, we discuss the changing political economy of agricultural price and trade policies. We follow with some concluding observations. Economic Growth and Structural Change The growth of gross domestic product (GDP) has accelerated steadily in Bangladesh since the early 1980s. The average growth rate rose from 3.2 percent in 1980­84 to 5.3 percent in 2000­04.1 The acceleration in per capita income growth has been even greater, mainly because the population growth rate has declined from 2.2 to 1.5 percent per year over the last 25 years. Several features of the pol- icy context during this period of growth are noteworthy, including the stable macroeconomic environment and the emphasis on the private sector as the engine of economic growth, on economic liberalization and outward orientation, and on agriculture and rural development. At only 2.8 percent in 1980­2004, the growth rate in the agricultural sector (including forestry and fishing) has been considerably slower than overall GDP growth. It was also uneven. It averaged only 1.7 percent a year during the first half of the 1990s, when growth in the crop subsector declined slightly. During the sec- ond half of the 1990s, growth in the crop subsector surged to an average 3.8 per- cent a year, while total agriculture grew at 4.8 percent a year. Growth in the fishing subsector was even more rapid in the 1990s, averaging 8.1 percent a year, but then slowed substantially to 1.7 percent after 2000 (Ahmed et al. 2007). Thus, as in most other countries, sustained economic growth in Bangladesh has been accompanied by a structural transformation that has involved a declin- ing share of agriculture despite positive agricultural GDP growth. The GDP share of agriculture declined from 32 percent in 1980­84 to 23 percent in 2000­04. About 7 percentage points of the growth throughout this period was accounted for by forestry and fishing. Meanwhile, the share of industry in GDP rose from 18 to 28 percent, and the share of services was steady at around 50 percent. Exports of ready-made garments have boomed in the past two decades, causing agriculture's share in total exports to fall from more than 33 percent in the 1970s and 1980s to less than 8 percent after 2000. Nonetheless, more than 60 percent of Bangladesh 307 the country's labor force continues to be employed in agriculture at least part time. Moreover, because agricultural production provides critical links for devel- opment in the rest of the economy, its performance has an important bearing on employment generation, food security, and poverty reduction. For these reasons, agricultural growth is still a development priority. Field crops and horticulture are the dominant activities in agriculture, accounting for four times as much agricultural value added as livestock activities since the early 1980s (Ahmed et al. 2007). Cropping is dominated by the produc- tion of cereals, especially rice, in which technological progress, supported by the development of irrigation infrastructure, has been the main engine of growth. Rice represented about 70 percent of the value added arising from crop produc- tion in 1973 and about 80 percent in 1999. The rapid expansion of rice produc- tion has been achieved in part at the expense of pulses and oilseeds, which remain important sources of protein and micronutrients. Apart from rice and wheat, the only crops showing rapid growth over the past two decades are maize, potatoes, and vegetables. Domestic demand factors have contributed to this structural change in the agricultural economy. They include income growth in aggregate, plus the acceler- ation in the growth in per capita income generated by the slowdown in popula- tion growth. Population growth was the dominant factor behind the growth in domestic demand for food in the 1970s, when per capita incomes were basically stagnant. In the 1990s, income growth and the varying income elasticity of demand for different food items became major factors behind the pattern of growth in domestic demand. Farmers have now started to respond to market sig- nals by reallocating resources to activities that are showing strong market growth (Hossain 2001). Agricultural exports In 1973, exports of agricultural commodities stood at slightly more than 80 per- cent of total exports. Their value grew at a modest trend rate of 4.2 percent in nominal terms and 1.4 percent in real terms over the subsequent three decades. Most of the growth was recorded by fishing (shrimps). In contrast, overall exports grew at a trend rate of nearly 11 percent (8.4 percent in real terms), which largely reflected rapid growth in ready-made garments. The share of ready-made gar- ments in total exports had risen to nearly 75 percent by 2004­05, while the share of agriculture in total exports had declined from 37 to 7 percent (Ahmed et al. 2007). The composition of agricultural exports has also undergone significant change during the last three decades. In the 1970s, the average share of exports of raw jute 308 Distortions to Agricultural Incentives in Asia in total agricultural exports was 70 percent, while the corresponding share of exports of tea and shrimps were 17 and 9 percent, respectively. However, exports of raw jute began facing stiff competition from synthetic substitutes, and the real value of raw jute exports declined by about two-thirds from the 1970s to the early 1990s before stabilizing, although the share of the exports in total agricultural export earnings continued to fall as exports of shrimps and vegetables expanded. By 2004, the share of jute exports was only 15 percent (and of tea exports, less than 3 percent) compared with 65 percent for shrimps and 6 percent for vegetables (from zero in the 1970s). The rising demand for vegetables by expatriate Bangladeshis and various policy measures, particularly the provision of cash incentives and subsidized freight charges, facilitated the rapid growth in vegetable exports. Agricultural imports Similar to the case of agricultural exports, the share of major agricultural imports in total imports has declined since the 1970s. The share of these imports--wheat, raw cotton, edible oil, rice, sugar, milk and cream, pulses, spices, oilseeds, and tobacco--declined from an average of 33 percent in the 1970s to 24 percent in the 1980s and only 16 percent in 2000­04 (Ahmed et al. 2007). The total real value of major agricultural imports nonetheless increased over this period by an average of 2.1 percent a year. This is more rapid than the increase in the value of agricultural exports (1.4 percent a year), but slower than the increase in the value of total imports (4.9 percent a year). In 2004, raw cotton, edible oil, wheat, rice, sugar, and milk powder accounted (in decreasing order) for most agricultural imports. However, the above figures cover recorded imports only. Studies have shown that illegal imports from India constitute nearly 20 percent of Bangladesh's total recorded imports (Bakht 1999). Of these illegal imports, two-thirds consist of only six agricultural goods: cattle (42 percent), sugar (7 percent), pulses (6 per- cent), milk powder (3 percent), spices (3 percent), and rice (2 percent). Historically, food aid has constituted an important component of the wheat imports of Bangladesh. During the 1980s and 1990s, food aid in wheat accounted for about two-thirds of total wheat imports; thus, average wheat food aid was about 1.0 million tons a year, and average total wheat imports were 1.5 million tons a year. Government commercial imports accounted for the remainder of total imports. Domestic wheat procurement, food aid, private commercial imports (since the trade liberalization in the early 1990s), and government com- mercial imports have contributed wheat to the public food distribution system. The amount of food aid has dropped substantially since 1999 following large gains in domestic grain production. Food aid never provided more than 0.5 percent of Bangladesh 309 rice supplies, however. Before 1992­93, the public sector accounted for all com- mercial imports of rice; but, in response to market incentives, private sector imports have increased since then to supplement domestic production and gov- ernment imports. In most years since 2000, raw cotton has surpassed rice and wheat, combined, as the major agricultural import. The growth in the real value of raw cotton imports has been especially rapid since the early 1990s, when the government started offering a 25 percent cash subsidy for exports of ready-made garments produced using domestic fabric, a policy that encouraged a surge in investment in textile spinning and composite mills (Bakht 2001a). Growth in the real value of raw cotton imports averaged 12 percent a year in the 1990s and 15 percent a year in 1999­2004. The real value of imports of edible oil rose by 3.2 percent a year from the early 1970s to 2004, so that the share of edible oil in total agricultural imports remained close to one-fifth over the period. Recorded sugar imports (not including sugar smuggled from India) increased even more rapidly, rising from 1.4 percent of the value of agricultural imports in the early 1970s to 7 percent in 1999­2004. Between 1973 and 2003, the domestic production of sugar increased by only about 100,000 metric tons, raising the output level from 90,000 tons in 1973 to around 190,000 tons in 2003. In contrast, domestic demand for sugar increased by nearly 700,000 tons during this period. As a result, imports of sugar increased from about 12,000 tons in 1973 to nearly 600,000 tons in 2003. Exchange Rate and Trade Policies The government followed inward-oriented, import-substituting trade policies beginning at independence in 1971 until major reforms were instituted in the 1980s and especially the 1990s. This section provides a brief history of the policies and the reforms during these years. The 1970s: Inward-oriented policies Faced with an imminent balance of payments crisis in 1971, the first independ- ence government continued the highly restrictive trade and exchange rate policies that had been in place when Bangladesh had been part of Pakistan. Quantitative restrictions on imports, high tariff rates, and a fixed, overvalued exchange rate were used to control imports, conserve scarce foreign exchange, and provide pro- tection for domestic industry. However, in line with the political philosophy of the government, the protectionist trade regime was maintained even after the possi- bility of a balance of payments crisis had decreased substantially in the mid-1970s 310 Distortions to Agricultural Incentives in Asia following a recovery in export earnings, an increase in foreign aid, a decline in world prices for grains and other commodities, and the 70 percent devaluation of the taka (Tk) in 1975 from Tk 8.9 to Tk 15.1 per U.S. dollar. Initially, under the fixed exchange rate system, all foreign exchange accrued to the government and was then allocated for competing uses through a discre- tionary and cumbersome mechanism of import licensing. However, in the early 1970s, in response to a crisis in foreign exchange, the government instituted a sys- tem, eventually known as the wage earner scheme, that allowed foreign exchange earned by Bangladeshis abroad--worker remittances--to be used to import par- ticular categories of goods. At the same time, under the export performance license system, certain exporters were given a specified proportion of their export earnings in the form of import entitlement certificates. The certificates could be used to import particular categories of goods, or they could be sold at a premium so that other traders could import such goods. There was thus a de facto dual exchange rate, whereby the export performance license premium reflected the market price of foreign exchange to some extent. In 1977, wage earners were allowed to sell their foreign exchange earnings directly in a market that came to be known as the secondary exchange market. A special foreign exchange rate, called the wage earner scheme rate, was available to them on this market. During the early years of restrictive trade policy, most agricultural commodi- ties were on lists of restricted or banned imports. These lists were aimed at pro- tecting producers from external competition by ensuring that producers received remunerative prices. There were also restrictions on agricultural exports, and some agricultural products were subject to export duties. The goal was to ensure that adequate stocks of agricultural commodities were available in the domestic market. However, fiscal incentives were also established to promote exports of nontraditional items. According to an export policy order issued in 1976, earnings from exports of nontraditional items were doubled from 15­20 to 30­40 percent.2 Duty drawback schemes were undertaken to promote exports of textile products manufactured using imported raw materials. The outcome of these autarkic trade and exchange rate policies was disap- pointing with regard to export development, the balance of payments situation, and overall economic growth, particularly considering the rapid growth achieved by East Asian economies that followed a more outward-oriented development strategy. Disenchanted with the import-substitution strategy (and encouraged by donor conditionalities), policy makers in Bangladesh and other South Asian countries began to adopt more open economic policies in the late 1970s. This shift was facilitated by a change in government in 1975; the new government had a more positive view on the development of a mixed economy and favored a more open trade regime. However, the export promotion measures that were Bangladesh 311 introduced in the late 1970s were limited in scope and were implemented only gradually. Overall trade policies continued to be inward looking, and the econ- omy, particularly the external sector, remained overregulated. The 1980s: Initial policy reforms The new government that came to power in March 1982 initiated a wide-ranging policy reform package, the New Industrial Policy, to liberalize the economy. A large number of nationalized companies, especially in the jute and textile subsec- tors, were privatized within a few years, and measures to increase foreign and domestic investment were adopted. However, only limited trade liberalization occurred under the New Industrial Policy or its successor, the Revised Industrial Policy, introduced in 1986. A cash compensatory scheme administered by Bangladesh Bank was also introduced in 1986, providing cash equal to 15 percent of free on board export values for exports manufactured using local fabrics.3 Ultimately, the progress in trade liberalization was slow in the 1980s, especially in the reduction in import tariffs (Bakht 2001b). However, some liberalization in agricultural exports did occur. Export duties on raw jute and tea were withdrawn in 1981, and the duties on exports of dried fruit, fresh fruit, oil cakes, coriander seed, dry chilies, dry ginger, black pepper, turmeric, tobacco, vegetables, and pota- toes were withdrawn in 1986. Restrictions on exports of jute seed, wheat, pulses, shrimps other than frozen, frogs (dead or alive) and frog legs, and onions were maintained even after 1995. Export restrictions on rice, wheat bran, and molasses were removed only in 1998. There were major reforms in exchange rate policy. In mid-1979, Bangladesh adopted a flexible policy, fixing the taka to a basket of the currencies of the coun- try's major trading partners.4 Then, in 1986, the export performance license system was simplified substantially through the introduction of an export per- formance benefit. Exporters could directly cash any benefit entitlements at their banks at the existing wage earner scheme rate. During the 1980s, these policies, particularly the export performance benefit, contributed to rapid growth in non- traditional exports. They also contributed to rapid expansion in the secondary (wage earner scheme) market. A wide range of agricultural commodities were favored by the export performance benefit incentives (Ahmed et al. 2007). How- ever, exports of raw jute were not included in this benefit scheme and thus suffered directly from the overvalued exchange rate. Meanwhile, imports financed at the official exchange rate were quickly reduced, and an increasing share of imports became subject to the secondary exchange market, including a portion of the foreign exchange received through commodity aid. By 1991, 41 percent of all imports were financed through this 312 Distortions to Agricultural Incentives in Asia means, while the share of the secondary exchange market in nonaid imports was nearly 70 percent. The enhanced role of the secondary exchange market helped narrow the gap between the official exchange rate and the wage earner scheme rate. The two rates were finally unified in 1992, and this marked the end of the export performance benefit arrangement (Rahman 1992). 1990s: Major trade liberalization After a decade of half-hearted attempts at trade liberalization, the democratic government that came to power in 1991 sought boldly to reform the trade regime. The reforms included tariff reductions and rationalization, the establishment of a less-complicated import tax structure, the gradual elimination of nontariff import restrictions, and promotion for exports through income tax exemptions, bonded warehousing, and flexible exchange rate management.5 Until the 1990s, the government relied heavily on quantitative restrictions to control imports, especially in agricultural commodities. About 37 percent of the tariff lines for agricultural products (21 percent of all products) were either banned or restricted in 1987 (World Bank 1994). By 1984, all quantitative restric- tions on agricultural products had been removed, and only 2 percent of the tariff lines of all products were still facing quantitative restrictions. In particular, private sector imports of rice and wheat were legalized in the early 1990s, ending the gov- ernment's monopoly on food grain imports. The ban on exports of fine quality rice (but not on ordinary coarse rice) was also lifted. Trade liberalization in the early 1990s brought tariff rates down sharply. Total protective import duty rates, both customs duties and para-tariff measures (sur- charges, license fees, regulatory duties, value added taxes, and supplementary duties), declined from 74 to 32 percent between 1991 and 1995 (measured accord- ing to the unweighted average of all tariff lines). Likewise, import tariffs and the total tax incidence on the imports of major agricultural commodities declined rapidly during the early 1990s (table 9.1).6 Refined edible oil, sugar, milk powder, and spices were subject to relatively high duty rates, while raw cotton, wheat, rape- seeds, and lentils enjoyed lower duty rates (table 9.2). Trade reforms have stalled in recent years, however. Although customs duties declined from 29 percent in 1995 to 19 percent in 2003, the effect of para-tariff measures rose sharply, mainly because of a steep increase in supplementary duties. Average total protective import duty rates have thus remained essentially unchanged since the mid-1990s. For some products that were already protected, including processed fruits, cement, soap, cotton shirts and sheets, selected ceramic and steel products, batteries, bicycles, and toys, total protection rates rose by more than 30 percent between 1997 and 2003 (World Bank 2004). Table 9.1. Tariff Rates on Imports, Bangladesh, 1991­2003 (unweighted average, %) All tariff lines Industry tariff lines Agricultural tariff lines Year Customs duties Para-tariffs Total rate Customs duties Para-tariffs Total rate Customs duties Para-tariffs Total rate 1991 71 3 74 70 3 73 77 0 77 1992 58 3 61 57 3 60 62 0 62 1993 43 2 46 43 3 46 46 0 45 1994 34 3 38 34 4 37 37 2 40 1995 29 3 32 28 3 32 30 2 32 1996 28 3 32 28 4 31 30 2 33 1997 27 6 33 27 6 33 29 5 35 1998 27 6 32 26 6 32 28 5 34 1990 22 7 29 22 7 29 25 5 30 2000 21 7 29 20 8 28 25 5 30 2001 21 8 29 20 8 29 25 8 33 2002 20 7 26 19 7 26 24 5 29 2003 19 10 29 18 9 27 23 17 40 Source: World Bank 2004. 313 314 Distortions to Agricultural Incentives in Asia Table 9.2. Total Taxes on Agricultural Commodity Imports, Bangladesh, 1992 and 2002 (percent) Commodity 1992 2002 Milk powder 72 74 Refined soybean oil 101 59 Refined palm oil 87 59 Sugar 135 47 Spices 80 39 Crude soybean oil 66 39 Tobacco -- 39 Rice 89 29 Lentils 20 14 Rapeseeds 20 14 Wheat 8 14 Raw cotton 8 7 Source: World Bank 2004. Note: -- no data are available. Impacts of Agricultural Price and Trade Policies on NRAs In this section, we consider the distortionary policies on several key crop prod- ucts: rice, wheat, sugar, and potatoes, plus jute and tea, the traditional export crops. Together, these products account for around three-quarters of the value of agricultural production at distorted prices (figure 9.1). In line with the methodol- ogy of our project (see Anderson et al. 2008 and appendix A), we estimate NRAs for the output of each of these products. Through careful comparisons of domes- tic prices and border prices or international reference prices (adjusted for quality differences, marketing margins, and the dual exchange rate system), these meas- ures capture the proportional extent to which government-imposed distortions create a gap between domestic prices and prices as they would be under free market conditions.7 Since it is not possible to understand the characteristics of agricultural development through a sectoral view alone, we use our project's methodology to estimate the effects of direct agricultural policy measures (including distortions in the foreign exchange market), but, for comparative eval- uation, we also generate estimates of the distortions in nonagricultural sectors. More specifically, we compute NRAs for the key products of farm producers. The NRAs include adjustments for direct interventions on farm inputs such as fertil- izer. We also generate NRAs for nonagricultural tradables for comparison with the Bangladesh 315 Figure 9.1. Commodity Shares in Agricultural Production, Bangladesh, 1971­2003 100 % 90 80 prices, 70 60 distorted 50 at 40 30 shares 20 10 value 0 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 year residual tea wheat potatoes sugar jute rice Source: Author calculations based on producer price and production data in FAOSTAT Database 2008. NRAs for agricultural tradables so as to calculate relative rates of assistance (RRAs) (see Anderson et al. 2008 and appendix A).8 Rice The impact of trade policy on agricultural incentives is largely determined by the price and trade policies for rice because of this commodity's predominance in the agricultural sector. Bangladesh has always been a net importer of rice, although only small quantities were imported in some years. From independence until 1992, private sector rice imports were banned, and government commercial imports accounted for almost all rice imports. (Rice was insignificant as an item in food aid.) Since 1993, when rice imports were liberalized, the private sector has accounted for most rice imports. Since 1980, the average annual volume of wheat imports has been about three times the corresponding volume of rice imports (about 1.6 million tons and 0.6 million tons, respectively) (table 9.3). The higher share of wheat in total food grain imports partly reflects inflows of food aid in wheat and partly reflects gov- ernment policy that tended to favor the use of lower-cost wheat in the public food distribution system. Although rice accounted for about one-third of the grain 316 Distortions to Agricultural Incentives in Asia Table 9.3. Production and Imports, Rice and Wheat, Bangladesh, 1973­2004 Indicator, year Rice Wheat Total Production, 1,000 tons 1973­78 12,255 259 12,514 1979­88 14,501 1,085 15,586 1989­98 18,230 1,305 19,534 1999­2004 24,822 1,476 26,298 Imports, 1,000 tons 1973­78 216 1,292 1,507 1979­88 308 1,655 1,963 1989­98 660 1,363 2,022 1999­2004 795 1,680 2,475 Imports/net availability,a % 1973­78 2.5 81.7 12.7 1979­88 1.9 60.6 11.1 1989­98 3.6 52.1 9.9 1999­2004 3.4 59.7 9.5 Sources: Ministry of Finance 2005; BBS, various. a. Net availability is estimated as production, less a 10 percent adjustment for seed, feed, and wastage, plus imports. Variations in public stocks are not counted. distributed through the system (the other two-thirds was wheat), rice distribution represented only 4 percent of the total net availability of rice from the 1970s through the 1990s. This is about the same as the corresponding share in East Pakistan in the 1950s and 1960s (see Ahmed, Nuruddin, Chowdhury, and Haggblade 2000). Total rice imports averaged about 3 percent of net rice availabil- ity over 1980­2004 (calculated without consideration of changes in public stocks). In contrast, wheat imports--largely food aid until the late 1990s--accounted for about two-thirds of total wheat availability during the period. In total, however, imports as a share of availability steadily declined, from an average of 17 percent in the 1980s to 14 percent in the 1990s and only 11 percent in 2000­04. From the early 1970s to 2005, the wholesale price of rice averaged close to the import parity price (calculated on the basis of the average price of rice imports). This masks wide fluctuations from year to year, however. This is so particularly during the period when the public sector held a monopoly on imports, and there was thus no direct link between international and domestic prices. The country experienced a famine in 1974 in the wake of the damage and destruction to transport infrastructure during the war in 1971 and the shortfall in rice production in 1974. The government lacked sufficient foreign exchange to Bangladesh 317 Figure 9.2. NRAs and Border Prices, Rice, Bangladesh, 1974­2004 600 120 100 ton 500 80 per 400 60 NRA, dollar 40 300 U.S. % 20 200 0 price, -20 100 border -40 0 -60 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 year border prices NRA Source: Ahmed et al. 2007. Note: The Pearson correlation coefficient between the two series shown is 0.41. make up the deficit by purchasing rice imports at the high prevailing international price.9 The domestic price for rice rose substantially in the 1974/75 season, but was still below the international price. After this crisis and the 61 percent devalua- tion in the taka relative to the U.S. dollar in May 1975, the domestic wholesale price for rice was double the import parity price, but it went below this level again during the subsequent six years (figure 9.2). Rice and wheat imports represented 10 percent of the total food grain supply, but, because wheat was not a close substitute for rice in domestic consumption, the effect of wheat imports (mainly food aid) on the market price for rice was likely to have been small. Nonetheless, in the absence of wheat imports, domestic rice prices would have been somewhat higher during this period.10 In the early 1980s, rapid increases in the domestic production of rice led to greater availability and lower real wholesale prices. International rice prices were falling even more rapidly, however, so that, over the five-year period 1983­87, the average domestic price was 21 percent above the import parity price. The govern- ment monopoly on imports thus had an effect on rice tariffs similar to the effect of an import tariff under a liberalized trade environment that favors net sellers of rice relative to net buyers. A surge in domestic rice production (mainly of winter rice [boro]) following major floods in 1987 and 1988 led to another drop in real 318 Distortions to Agricultural Incentives in Asia wholesale rice prices in 1989, and this brought domestic prices close to par with border prices and eliminated the implicit protection for producers. This price-sta- bilizing role of government has tended to generate a negative correlation between NRAs and border prices, the extent of which is evident in figure 9.2. Trade liberalization in the early 1990s created a possible direct link between domestic and border prices for rice, particularly during periods following poor domestic rice harvests. Throughout the 1980s and early 1990s, Thailand was the major source of rice imports to Bangladesh. However, the 1994 liberalization that permitted private sector imports in Bangladesh coincided with rice trade liberal- ization and the buildup of public rice stocks in India, which made private sector rice imports from India legal and feasible.11 Because of lower transport costs, reduced delivery times (for private sector imports), and the possibility of smaller contracts for the delivery of imports by truck, India rapidly replaced Thailand as the major source of rice imports in Bangladesh in the mid-1990s. Large-scale rice imports from India supplemented domestic supplies during crop shortfalls in Bangladesh in 1994, 1997, and 1998.12 In 1997 and 1998, 92 percent of rice imports came from India. However, during periods following normal or above- average rice harvests (most of 1996 and 1997, in 2000, and in the first half of 2001), domestic rice prices in Bangladesh were below import parity levels, elimi- nating any incentives for private sector imports (figure 9.2).13 Average domestic wholesale prices were 5 and 12 percent below average import parity prices in 1990­94 and 1995­99, respectively, although there were few trade barriers during these years. The negative NRAs reflect periods when rice came close to being a nontradable because of good domestic harvests. Beginning in 2000, the government of India adopted increasingly aggressive measures to promote exports to reduce a massive public stock buildup. The meas- ures included subsidies for rice exports that involved the provision of grain from government stocks to exporters at prices that were below cost.14 Because domestic rice prices in Bangladesh were approximately equal to the full-cost import parity prices, including taxes, for below-poverty-line rice made in India, small amounts of low-quality rice were imported into Bangladesh in 2000.15 However, when the Indian government lowered its above-poverty-line sales price for fine rice from Rs 11.3 per kilogram to only Rs 8.3 per kilogram in July 2001, the government of Bangladesh increased rice import tariffs and taxes from 5 to 37.5 percent.16 This raised the below-poverty-line import parity price, plus taxes, to 33 percent above the domestic price in Bangladesh, thereby essentially cutting off the incentive for private sector rice trade with India (figure 9.3).17 However, in mid-2002, the domestic price again rose toward the import parity price, including taxes, leading to relatively large-scale private sector trade with India once more. Until early 2006, domestic prices were at or near below-poverty-line import parity prices, Bangladesh 319 Figure 9.3. Prices and Private Sector Imports, Rice, Bangladesh, 1997­2007 35 400 30 350 private 300 25 sector 250 20 200 imports, taka/kilogram 15 150 price,10 tons 100 5 50 0 0 Jan-97Jul-97Jan-98Jul-98Jan-99Jul-99Jan-00Jul-00Jan-01Jul-01Jan-02Jul-02Jan-03Jul-03Jan-04Jul-04Jan-05Jul-05Jan-06Jul-06Jan-07Jul-07 month, year private sector imports import parity, Bangkok exports Dhaka wholesale price import parity, New Delhi exports below-poverty-line import parity price Source: Ahmed et al. 2007. Note: Import parity prices include taxes. including taxes, suggesting that the former were essentially being determined by the latter during this period.18 In 2002­04, the average wholesale price in Bangladesh was only 1 percent below the below-poverty-line import parity price, including taxes. Import tariffs raised the domestic price relative to the import parity price, without taxes, for subsidized Indian below-poverty-line rice by an average of 10 percent. However, the domestic price in Bangladesh was 28 percent below the import parity price based on the wholesale price in New Delhi and 15 percent below the import parity price based on Bangkok exports. In the absence of these subsidized imports, the net domestic supply of rice, without government intervention, would have been 5 percent less over 2002­04, and the domestic rice price would have been about 15­25 percent higher. The lower value assumes an own-price elasticity of rice demand of 0.3, and the higher value assumes an own-price elasticity of rice demand of 0.2. Thus, under the first assumption, the domestic price would have risen close to the import parity price based on Bangkok exports, and net imports would have been near zero. 320 Distortions to Agricultural Incentives in Asia Wheat Wheat was a minor crop in Bangladesh in the 1960s, prior to independence. Wheat production increased rapidly thereafter, from an average of about 0.1 mil- lion tons a year in 1969­74 to more than 1.8 million tons a year in 1997­99. This rise was associated with a sevenfold expansion in the area cultivated in wheat and a twofold expansion in wheat yields per hectare. Wheat production growth was especially rapid in the 1970s, increasing by an average 37 percent a year as the area under wheat expanded by 19 percent a year, and yields per hectare rose by 15 per- cent a year. In 1998­2003, the area under wheat declined from a peak of 967,000 hectares to 704,000 hectares, while the area planted in maize, potatoes, and winter rice expanded (Dorosh 2006). Because of the government monopoly on external trade, there was no explicit link between the international price of wheat and the domestic price from the 1970s to the early 1990s. Because food aid and government commercial imports accounted for over half of the total supply, government policy on the net distribu- tion of wheat (there was limited domestic procurement of wheat) was the domi- nant factor determining the domestic market price. Partly because of food aid conditionality among donors stipulating that food aid should not generate price disincentives for domestic production, the average domestic price on wheat was close to the import parity price throughout this period. After the devaluation in 1975, the estimated NRAs on output averaged only 3 percent until 1993 (table 9.4). After the liberalization of wheat imports in 1992, the import parity price of wheat provided a price ceiling for wheat in the same way the import parity price of rice accomplished this for rice. In most years, domestic prices were close to this ceiling, so that the average NRA on output was less than 4 percent after 1994. Only during a period when international wheat prices were high (around 1996) were domestic prices substantially below border prices (NRAs of 14 percent). From the mid-1990s until 2000, most commercial wheat imports involved high-gluten wheat for baking that was imported from major international wheat exporters such as Australia, the European Union, and the United States. This wheat accounted for about 10 percent of total wheat use in Bangladesh, and it did not directly compete with soft, lower-gluten domestic wheat, which is used mostly for traditional breads (for example, roti and chapati). However, in 2000 and in sev- eral years since then, subsidized wheat exports from India were found on markets in Bangladesh. Because of the rapid increases in rice production and rice supply and the lower real price of rice, which tended to reduce the demand for wheat, the significant food aid beginning in 1999, if it had been maintained, would likely have reduced Table 9.4. NRAs for Covered Agricultural Products, Bangladesh, 1974­2004 (percent) Indicator, product 1974 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Exportablesa 28.7 34.6 26.2 32.4 33.0 9.9 33.2 Jute 30.0 37.1 29.3 35.4 38.4 5.6 38.7 Tea 1.4 14.5 10.7 19.9 11.9 20.5 20.4 Import-competing productsa 20.6 6.5 1.9 24.4 0.1 7.9 6.1 Riceb 25.7 6.5 5.2 20.4 5.3 12.0 2.6 Wheat 38.9 30.3 5.8 11.3 4.5 2.6 0.3 Sugar 73.7 92.1 137.0 436.0 166.1 138.8 223.9 Nontradablesa Potatoes 1.3 1.5 1.3 1.7 2.2 2.7 1.8 All covered productsa 20.8 2.8 3.8 16.8 2.2 7.6 3.9 Dispersion, covered productsc 52.1 71.4 67.6 190.7 77.5 67.9 101.2 Coverage, at undistorted prices 77 77 78 72 72 70 72 Source: Ahmed et al. 2007. a. Weighted averages; the weights are based on the unassisted value of production. b. Calculated based on the average cost of rice imports. c. Dispersion is a simple five-year average of the annual standard deviation around the weighted mean of the NRAs of covered products. 321 322 Distortions to Agricultural Incentives in Asia the domestic price of wheat below the import parity price (del Ninno and Dorosh 2003). However, donors substantially reduced food aid after 2001. There were three reasons for this reduction: the government had achieved its domestic food production target of 454 grams per person per day, thus eliminating the notional food gap; the European Union was reducing its use of in-kind food aid; and the United States and other countries were shifting their deliveries of food aid resources to other parts of the world. Nonetheless, wheat production has declined since 2000. This has largely been the result of growing competition from maize, including hybrid varieties that have been profitable and have been satisfying a domestic market for poultry feed. Jute After the onset of the boom in the production of ready-made garments in Bangladesh in the early 1980s, jute lost its prominent share in total export earn- ings. Nonetheless, it is still an important farm export commodity. Bangladesh remains the leading exporter of raw jute in the world, and a significant number of farm households continue to depend on jute cultivation for their livelihoods. International jute prices have been declining steadily, however, and the world demand for jute has been shrinking in the face of synthetic substitutes. The nominal U.S. dollar export price of jute fell by an average 1.5 percent a year from 1973 to 2004. The decline in real prices has been an even steeper 4.2 percent per year (measured using the U.S. wholesale price index as a deflator for the dollar price series).As a result, the real 2004­05 dollar price of jute fell by two-thirds between the late 1970s and the early 2000s. Because the quantity of raw jute exports remained rather flat over the period, export earnings fell dramatically (Ahmed et al. 2007). The domestic price of jute has been consistently below the export parity price partly because of an effort to encourage the domestic processing of jute products for export.19 In the 1970s and 1980s, the average domestic wholesale price of jute was more than 30 percent below the export parity price.20 Following the trade lib- eralization of the early 1990s, the NRAs on jute output averaged only 6 percent. However, after the closing of public sector jute mills, the domestic price of raw jute again fell significantly below the export parity price, perhaps reflecting the disruption in marketing channels and the greater differential between the quality of raw jute for export and jute for the domestic market (table 9.4). Tea Like jute, tea was once a major export of East Pakistan and then Bangladesh. Between 1973 and 1983, the value of tea exports rose in nominal dollar terms Bangladesh 323 mainly because of an increase in the international price of tea. However, from the peak value of exports, at US$69 million in 1983, export earnings from tea declined steadily to only US$16 million in 2004. One factor in the decline in exports was the rise in the domestic demand for tea. During the last three decades, tea produc- tion has expanded at a trend rate of 2 percent, while domestic tea consumption increased at a trend rate of 7.5 percent, leaving a smaller exportable surplus. The domestic price of tea, which is based on the auction at Chittagong, has been consistently 10 to 20 percent below the export parity price, which is equiva- lent to the average export price in Chittagong. This may reflect handling costs at the port. Rahman (1994) uses a border price based on the London auction price, discounted by 30 percent, less marketing and processing costs to cover the trans- fers to Chittagong port from the tea gardens in Sylhet in northeastern Bangladesh. He finds little difference between the domestic price and the border price in most years between 1974 and 1987 except when the border price spiked in 1976, 1983, and 1984. The gaps between the average export price and the Chittagong auction price or the London auction price may largely reflect quality differences rather than trade and domestic pricing policy distortions. Sugar Bangladesh produces less than 20 percent of the sugar it consumes. The remain- der is provided by official sugar imports (nearly 50 percent of consumption) and sugar smuggled from neighboring India.21 The country's sugar industry is highly protected through import tariffs. Imports have been controlled through licensing (before 1992), a government monopoly (1992­2001), and a combination of parastatals and private traders (since June 2002). As a result, the domestic price of sugar is substantially higher than the world price. Farmers do not reap all the ben- efits of this policy, however, since sugarcane prices are fixed by the government, and sugar mills retain a monopsony on sugarcane purchases within designated sugarcane areas near each mill.22 Official trade has generally faced high import tariffs. For example, tariffs on sugar imports were 135 percent in 1992 before major trade liberalization, but still 47 percent in 2002 after the liberalization. These high tariffs on sugar have greatly reduced imports of sugar through official channels, but have also provided major incentives for sugar smuggling. Our NRA estimates for output vary substantially over time largely because of fluctuations in the world price of sugar, but also because milling, transport, and marketing costs are subtracted from the border price for sugar to determine the sugarcane price at the farmgate. Our NRAs on output average more than 150 per- cent for the 1980s and 1990s and even more since 2000 (table 9.4). 324 Distortions to Agricultural Incentives in Asia Distortions in input prices Modern agricultural technology was introduced in East Pakistan in the early 1960s. There was eventually heavy public sector involvement in the procurement and distribution of modern agricultural inputs and in investments for the devel- opment of water resources. A parastatal, the East Pakistan Agricultural Develop- ment Corporation, later known as the Bangladesh Agricultural Development Corporation, was established in 1963. The corporation held a virtual monopoly over the procurement and distribution of fertilizers, seeds, pesticides, and small irrigation equipment, although it had to conform to pricing policies and related policies that the government formulated from time to time.23 Major reforms in the markets for fertilizer and irrigation equipment were launched in the late 1970s (Ahmed et al. 2007). Under the new marketing system established in 1978, private sector trade in fertilizer was liberalized, leading to a substantial expansion in the number of wholesalers and retailers operating in the fertilizer market.24 The share of private trade climbed to 75 percent in 1989 and nearly 100 percent in 1992, when the ban on private sector fertilizer imports was removed and the deregulation of fertilizer marketing was completed. Between December 1994 and March 1995, however, a serious shortage in the supply of urea led to a crisis in the market. The government partially reversed the reform process by initiating controls on wholesale markets, regulating pricing at the factory gate, and imposing restrictions on domestic traders who were selling outside the dis- tricts in which they had been registered (Ahmed 2001).25 Subsidies for the domestic production of nitrogenous fertilizer have been responsible for the greatest distortion in agricultural input prices. By 1983, fertil- izer prices at the farmgate had been deregulated throughout the country and direct subsidies to farmers for the production of urea had been eliminated. How- ever, subsidies continued through 1991 for imports of other fertilizers--namely, triple superphosphate and muriate of potash--that were designed to improve the balance of chemical nutrients. Subsidies for imported fertilizers such as these were reintroduced in January 2005 at the rate of 35 percent. Government controls on urea and on the volume of urea imports and exports have helped maintain the domestic wholesale price for urea consistently below the import parity border price and generally above the export parity price, despite the liberalization in marketing.26 The average domestic price for urea was 37­50 percent below the import parity price in the 1970s and 1980s and 50 per- cent below the import parity price in the 1990s and 2000s. (The average domes- tic price for urea was about 40 percent above the average export parity price from 1990­91 to 2004­05.) However, the average domestic price of triple superphos- phate has been only 18 percent below the import parity price since 1990, though Bangladesh 325 it had been 47 percent below the import parity price in the 1970s and 1980s (Ahmed et al. 2007). Given that the cost of fertilizer accounts for an average of only 5 percent of the total cost of paddy and even less for most other crops, the fertilizer subsidy has not had a major effect on the overall NRAs in agriculture in recent years. Indeed, it has added only 1 or 2 percent during most years. This also applies to the main non- tradable food, potatoes, which would otherwise show NRAs of zero. The total assistance to agriculture For the agricultural sector as a whole, including both tradable and nontradable goods, average NRAs since the mid-1970s have been low. Since 1975, the five-year average NRAs for the six products we cover have ranged from 8 percent in the mid-1990s, when international prices were high, to 17 percent in the later 1980s, when international prices were low (table 9.4). Three key factors have driven this result. First, since rice accounts for two-thirds of covered product output, the NRAs for rice largely determine the total NRAs for tradable agriculture. The prod- uct exhibiting the greatest price distortions, sugarcane, accounts for only 1 or 2 percent of the total value of domestic agricultural production in most years and so has little influence on the sector's average NRAs. Second, the implicit taxation of agricultural exports (jute and tea) partially off- sets the protection for importables in our total NRA estimates. However, jute and tea account for only a small weight in the average NRA (less than 5 percent since the late 1980s). Third, while the share of potatoes, the nontradable good shown in table 9.4, has a somewhat larger weight in the average, up to 7 percent in recent years, the only measured assistance involved arises from the assistance for fertilizer, and, so, the NRA for potatoes is small. Our NRA estimates suggest that, relative to import-competing agricultural products, agricultural exports are discouraged. This is illustrated in figure 9.4. Our estimates also suggest that the dispersion of the NRAs for our covered prod- ucts around the mean value each year is wide and has not declined over the past 30 years. One measure of this phenomenon, shown near the bottom of table 9.4, is the standard deviation in these NRAs: it was around 70 percent or more. This lack of convergence in NRAs implies that the allocation of land and other farm resources among our covered industries continues to be inefficient. Virtually all noncovered products, including fruits, vegetables, and meat products, receive little assistance; we therefore assume that the corresponding NRAs are zero. We also assume that they are nontradable over the period we study. Including these 326 Distortions to Agricultural Incentives in Asia Figure 9.4. NRAs for Exportable, Import-Competing, and All Agricultural Products, Bangladesh, 1974­2004 120 100 80 60 40 % 20 NRA, 0 -20 -40 -60 -80 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 year import-competing total exportables Source: Ahmed et al. 2007. goods, which account for about one-quarter of the total value of agricultural produc- tion, the weighted average NRAs for total agriculture are closer to zero relative to the weighted average of the covered products alone (upper portion of table 9.5). The assistance provided by government policies to nonagricultural tradable sectors is also important for the competitiveness of agriculture within the econ- omy and the ability of agriculture to contribute to the economy. The effect of import tariffs and import quotas on the domestic prices of nonfarm import-com- peting goods may be expressed as an implicit tariff rate. This rate is defined as the ratio of domestic prices (measured at the border) to import prices. In the absence of detailed data on domestic and import prices, we have calculated estimates of this implicit tariff. Our calculations provide NRAs for the import-competing parts of the nonfarm sector. We combine these with an assumed NRA of zero for the exportable part of the same sector to generate production-weighted average NRAs for nonagricultural tradables. Though this method may be crude, it pro- vides a reasonable measure that may be compared with the NRAs for tradable agriculture so as to derive RRAs. The RRAs indicate the extent to which the prices received by farmers are depressed relative to the prices faced by the producers of other tradables in the economy (table 9.5, note d.). Table 9.5. NRAs in Agriculture Relative to Nonagricultural Industries, Bangladesh, 1974­2004 (percent) Indicator 1974 1975­79 1980­84 1985­89 1990­94 1995­99 2000­04 Covered productsa 20.8 2.8 3.8 16.8 2.2 7.6 3.9 Noncovered products 0.0 0.0 0.0 0.0 0.0 0.0 0.0 All agricultural productsa 16.0 1.4 3.3 11.7 1.5 5.2 2.7 Non-product-specific assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total agricultural NRAb 16.0 1.4 3.3 11.7 1.5 5.2 2.7 Trade bias indexc 0.10 0.30 0.23 0.45 0.33 0.00 0.37 NRA, all agricultural tradables 21.6 3.1 3.9 17.5 2.4 8.0 4.0 NRA, all nonagricultural tradables 45.9 28.4 22.4 28.5 33.3 29.0 23.4 RRAd 46.3 19.7 21.5 8.6 26.7 28.6 15.8 Source: Ahmed et al. 2007. a. Including product-specific input subsidies. b. Including product-specific input subsidies and non-product-specific assistance. Represents total assistance to primary factors and intermediate inputs, divided by the total value of primary agriculture production at undistorted prices. c. The trade bias index is defined as (1 NRAagx 100) (1 NRAagm 100) 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector, respectively. d. The RRA is defined as 100*[(100 NRAagt) (100 NRAnonagt) 1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. 327 328 Distortions to Agricultural Incentives in Asia Figure 9.5. NRAs for Agricultural and Nonagricultural Tradables and the RRA, Bangladesh, 1974­2004 100 80 60 % 40 RRA, 20 and 0 NRA -20 -40 -60 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 year NRA, agricultural RRA NRA, nonagricultural tradables tradables Source: Ahmed et al. 2007. Note: For the definition of the RRA, see table 9.5, note d. In table 9.5, we summarize the RRA findings on the basis of annual estimates reported in Ahmed et al. (2007). These estimates reveal that, although the NRAs for agricultural tradables were positive in some years, they were always well below the NRAs for nonagricultural tradables. Hence, the RRA estimates are quite nega- tive; they suggest that the relative prices of farm products have been depressed by more than one-fifth since independence. The value of the RRAs in the first half of the present decade is considerably higher than the value of the RRAs for the 1990s, however (at an average 16 percent compared with 27 percent). Additional reform in the nonfarm sectors will be needed to eliminate the distortions in resource allocation among the sectors that produce tradables (figure 9.5). The Political Economy of Agricultural Policies There was a progressive shift in government agricultural policies toward privatiza- tion, deregulation, and a reduction in input subsidies. The shift began in the mid- 1970s and continued in stages up to the early 1990s. There are still government Bangladesh 329 controls on fertilizers, sugar production, and the wholesale trade, but these are of relatively minor economic significance. The successive government administrations that formulated and implemented these policies, like all governments, balanced a variety of objectives against a range of constraints. In the policy-making process, pressures and counterpres- sures are exerted on governments by various interest groups, and the responses of the governments are usually conditioned by interests held in common by the pres- sure groups and government leaders. Sometimes, if effective pressure groups have not emerged or if they may be successfully set against one another, the govern- ment might then pursue its own agenda, which may be predatory, altruistic, or a blend of the two (Grindle 1991). Even if reform proposals are sound on the economics, the reforms will not be successful and sustainable unless they find acceptance according to minimum social and political criteria. If they do not find such acceptance, the government may be forced to backtrack and adopt policies that are less satisfactory on economic grounds.27 In Bangladesh, several factors have been especially important in determining the influence of various interest groups on agricultural policies and reforms since independence. These include the relative political strengths of farmers versus urban groups, academic and political views on socialism and capitalism, debates within government ministries, and the reach of donors. One of the most important facets of the political economy is the relative weak- ness of farmers and the rural poor in lobbying the government. These groups include a large number of households, but their geographical dispersion, internal differences, ideological orientations, and poor resource base render them largely ineffective politically. In contrast, the working class in the urban formal sector, the bureaucracy as a pressure group within the government apparatus, and private entrepreneurs constitute organized and powerful interest groups that few govern- ments can afford to antagonize.28 The weak political position of farmers and smallholders explains why conflicts of interest between agriculture and industry have consistently been resolved in favor of industry. Thus, for example, in the case of agricultural-based industrial raw materials such as jute, the policy has focused on keeping the price of the input low so that the relevant industrial product is competitive. Export taxes and restrictions on agricultural commodity exports contributed to the discrimination against agriculture. The weak representation of the interests of the peasantry within the major political parties and the predominance of trading and industrial interests led to an excessively protected economic regime and shaped the political economy so that government policies continued to discriminate against agricul- ture. Even in the case of policy measures such as input subsidies, farmers failed to derive much benefit because the subsidies were largely usurped by rent-seeking 330 Distortions to Agricultural Incentives in Asia public officials who were in collusion with middlemen. Any residual benefits to farmers were often more than offset by depressed farm output prices. Another important factor shaping agricultural and economic policies more generally is the wide divergence of views among social scientists and other profes- sionals on the issue of reform. The views of some critics of market mechanisms were initially shaped by colonialism, which these critics consider a form of exploitation by world capitalism. These critics once pointed to the apparent suc- cesses of China and the Soviet Union in economic transformation and to certain blatant examples of market failure, such as the Bengal famine in 1943. The strides taken in China toward the market economy and the disintegration of the Soviet Union have, more recently, disillusioned many of these critics regarding the virtues of central planning. Nonetheless, an articulate antimarket lobby still per- sists among intellectuals that effectively counters the sweeping and sometimes simplistic claims for the market made by votaries. The policy process is also complicated by the fact that policy makers and implementers often hold divergent views on the appropriateness of particular reform measures. Sometimes, this occurs because of differences in the social and political orientation of these actors. Rivalry among ministries and the unwilling- ness of the bureaucracy to relinquish the levers of control even in cases in which political leaders are committed to deregulation may also lead to such outcomes. For example, a ministry of commerce may obstruct liberalization measures pro- posed by a ministry of finance. Finally, donors have had a major effect on government policy in Bangladesh in part because of the importance of foreign aid in the development budget and in balance of payments support. Within agriculture, donor support in the funding for agricultural research, rural infrastructure, and food aid was especially impor- tant in the 1970s and 1980s, and this contributed to the weight of donor views in the policy process. In particular, the Asian Development Bank, the United States Agency for International Development, and the World Bank exerted substantial influence during the formulation and implementation of agricultural policy by tying program loans and import credits to the policy reform agenda. Major Reforms in Agricultural Pricing The major reforms in agricultural policy in the 1980s were swayed by the policy perspectives of the dominant donor agencies as outlined in papers on agricultural and food policy and pricing prepared by the World Bank and the United States Agency for International Development. The World Bank (1979) argued that the net effect of policy interventions in developing countries was to discourage agricultural production, that higher price incentives might be an important Bangladesh 331 instrument in increasing food production, and that input subsidies had proved an inefficient means of protecting agriculture. The World Bank also highlighted the complexities in the effects of food prices on various groups and advocated a grad- ual rise in food prices to avoid the hardships resulting from abrupt price increases. The United States Agency for International Development (USAID 1982) likewise argued that government interventions in agricultural markets tended to reduce efficiency in the allocation of resources, inhibited gains in productivity, and often did little for the groups they were supposed to help. The U.S. aid agency advocated policies that did not suppress producer prices, that did not attempt to regulate consumer access to food, and that led to a speedy withdrawal of subsidies for agri- cultural inputs, including credit. Government interventions were to be limited to agricultural research, the construction of large-scale irrigation systems, and spe- cial feeding programs to combat malnutrition. Thus, donors generally advocated a market-oriented agricultural policy, though they also suggested that a central role should be reserved for output price supports to promote self-sufficiency in food grains.29 This objective was ulti- mately achieved in 2000 (Osmani and Quasem 1990).30 Reforms in agricultural input markets, particularly those related to fertilizer prices, generally faced more opposition. In 1973, the First Five-Year Plan of Bangladesh (1973­78) underscored the need for the eventual removal of agricul- tural input subsidies and emphasized that the inputs should be sold at a profit. Indeed, urea prices were raised substantially in 1973. Sections of the bureaucracy were opposed to the elimination of input subsidies and to the privatization of the distribution of fertilizer and irrigation equipment. This resistance was partly caused by the bureaucracy's reluctance to accept a curtailment in its power over market controls and distribution. A misplaced concern about equity also con- tributed to the opposition of bureaucrats to the proposed reforms. Donor pressure to carry out the reforms swiftly played a major role in quelling this opposition. Fertilizer markets were liberalized again in the early 1990s. The country wit- nessed a major crisis in the fertilizer market in 1994­95. This led to a partial rever- sal of the reform in the fertilizer market. Indeed, farmers agitated so militantly that there were incidents involving police gunfire and several deaths. This was in sharp contrast to the 1974 famine, which caused little uproar in rural Bangladesh, and to the fertilizer shortage in 1993 when farmers were paying high fertilizer prices with- out much protest (Abdullah 1996). The unmet expectations about the lower prices that were supposed to be generated by policy reforms and a perception that there had been injustices may explain the outrage of farmers in the mid-1990s.31 The 1974 famine had a different dynamic in the sense that it affected mainly landless laborers. In contrast, the fertilizer shortage in 1993 mainly affected mid- level farmers, who are known to be more militant and among whom marketed 332 Distortions to Agricultural Incentives in Asia rice surpluses are not usually sufficiently large to offset higher urea prices (Wolf 1971). Because national elections were pending, the political parties in opposition also tried to reap dividends from the crisis by mobilizing the peasantry. Not all reforms in agricultural input markets have been opposed by farmers. Some policy reforms seem to have benefited a majority of farmers. For example, the elimination of the standards requirement in irrigation equipment has been popular. This policy faced opposition from public sector employees, particularly those active in exercising government market controls and government authority over imports (especially the Bangladesh Agricultural Development Corporation). However, this opposition did not prove to be a major roadblock. The issue was never politicized, and the government and its development partners handled the matter tactfully, but also with firmness (Abdullah and Shahabuddin 1993). Concluding Observations Agriculture has undergone major structural changes and achieved major suc- cesses in Bangladesh over the last four decades. Despite the many problems and constraints, a quiet agricultural revolution has taken place that has enabled the country to achieve its national food security targets in the production of food grains. Agriculture continues to evolve in response to numerous factors, including natural calamities, sociopolitical changes, population growth, urbanization, new technology, opportunities in the rural nonfarm sector, and commercialization. Government macroeconomic, trade, and agricultural pricing policies have played a major role in shaping price incentives in production and consumption and will continue to be important determinants of agricultural growth. For over 30 years, a central objective of government agricultural policy was self-sufficiency in food grains. To achieve this objective, the government attempted to maintain sufficient incentives for the expansion of domestic rice and wheat production through the maintenance of remunerative output prices and fertilizer subsidies. Investments in agricultural research and technology that per- mitted large gains in productivity and a substantial increase in irrigation in the area under rice cultivation were also crucial. Meanwhile, government policy was also designed to protect poor consumers through the subsidized sale of rice (until the early 1990s) and extensive safety nets involving food for work programs and food transfers, particularly wheat. Although trade liberalization has faced substantial opposition, the govern- ment nonetheless undertook major reforms in trade policy, including reducing tariffs on industrial products in the 1980s and the early 1990s and liberalizing private sector trade in rice and wheat in the 1990s. As a result, the domestic out- put price for wheat and rice (the main agricultural products in value) has been Bangladesh 333 close to the border price in most years since the early 1990s. Thus, price distor- tions in agriculture have averaged less than 5 percent of the value of domestic pro- duction since 1990 despite the ongoing price distortions on a few products (notably, sugarcane) and inputs (chemical fertilizers). The country has reaped great benefit from trade liberalization through enhanced food security because private sector imports have helped stabilize markets following significant produc- tion shortfalls. Keeping the domestic prices of most agricultural commodities close to the respective border prices has also generated overall efficiency gains in the agricultural sector. Reducing the remaining disincentives for agricultural production--caused by the protection for nonagricultural producers--will be a necessary part of any future strategy aimed at agricultural growth and rural poverty reduction. Even a liberalized trade policy would not guarantee higher incomes among farmers. For example, in the early part of this decade, the upward trend in the ratio of fertilizer prices to paddy prices that was partly driven by movements in world prices even- tually reduced the price incentive for paddy production and contributed to lower returns to farmers. In 2007­08, the world prices for fertilizer and rice rose sub- stantially and, combined with the limits imposed by India on rice exports to Bangladesh, contributed to large increases in the domestic prices for fertilizer and rice in Bangladesh in early 2008. Policies aimed at increasing production and sta- bilizing prices need not rely mainly on price subsidies or substantial increases in public stocks, however. Indeed, productivity-enhancing investments in agricul- tural research and extension, improvements in postharvest management and agroprocessing, and investments in market infrastructure can complement agri- cultural price and trade policies and enable rapid agricultural growth and higher farmer incomes even in a context of shifting world prices. Notes 1. Throughout this chapter, crop or fiscal years are indicated by the second of the two calendar years. For example, 2004 here refers to the year 2003/04 in official statistics. 2. By 1995, exporters were also enjoying income tax exemptions on 50 percent of their export earnings. 3. Other measures included the export policy order issued in 1986, which provided special induce- ment and promotional freight rates by the national flag carriers, Biman Bangladesh Airlines and Bangladesh Shipping Corporation, for exports of fruits, vegetables, and ornamental plants. 4. Initially, the reference currency was the U.K. pound sterling; but this was changed to the U.S. dollar in early 1983 because of the large weight of the United States in the total trade of Bangladesh. 5. In 1996, the government accepted the conditions of article VIII of the Articles of Agreement of the International Monetary Fund by making the taka fully convertible for international current account transactions. Under this arrangement, exporters may freely utilize their export earnings to pay for imports. 334 Distortions to Agricultural Incentives in Asia 6. For details on the duty structure affecting imports of major agricultural commodities in 1991­2003, see Ahmed et al. 2007. 7. Our NRAs for tradables include an estimate of the trade tax effect of the overvalued exchange rate. For this estimate, we rely on the black market exchange rate premium (see Global Development Network Growth Database). We also assume that only half of the foreign exchange rate earnings of exporters is sold to the government at the official rate. For details on this methodology, see Anderson et al. (2008) and appendix A. 8. Throughout this section, the term agricultural sector refers only to crop and livestock production and excludes fishing and forestry. 9. See Ravallion (1990) and del Ninno, Dorosh, and Islam (2002) for detailed discussions of mar- kets and government policy during this period. 10. To analyze this policy fully would require a multimarket model (see Dorosh and Haggblade 1997). 11. Other factors also contributed to the increase in rice exports from India, including a 27 percent depreciation of the Indian rupee in real terms (see Dorosh 2001). 12. Several successive rice harvests were poor in Bangladesh: wet season rice (aman) in 1994, 1995, and 1997; winter rice in 1995; and severe losses in autumn (aus) rice and wet season rice following the mid-1998 flood. See Dorosh et al. (2004) for a discussion of government trade and pricing policies in food grains following the 1998 flood. 13. Small amounts of nonparboiled rice were imported in 2000, mainly from Vietnam. Higher- quality (noncoarse) rice imports likely accounted for much of the rest (Dorosh 2001). 14. For example, in the targeted public distribution system in India, wheat and rice were distributed at a special, highly subsidized rate to consumers certified as living below the poverty line; in 2000, state trading parastatals in India could buy wheat at these below-poverty-line prices and then export the wheat (USDA 2001; see also Dorosh and Shahabuddin 2005 and chapter 10, on India, in this volume). 15. Bangladesh imported 281,000 tons of rice across the land borders with India during 2000. This was about the same as the 286,000 tons imported in 1999. A breakdown of the rice by quality is not available, but a 9 percent decline in the average price of imports from India--from Tk 12.1 per kilo- gram in 1999 to Tk 11.1 per kilogram in 2000--suggests that the average quality of imports may have declined over the two years. 16. The total tariffs and fees were raised from 10.5 to 43.0 percent if one includes an advanced income tax of 3 percent and a license fee of 2.5 percent. 17. The increase in import taxes raised the above-poverty-line import parity price, including taxes, to 71 percent above the domestic price. 18. Our NRA estimates on output using the average cost of imports suggest that these NRAs were around zero, perhaps because shipments of high-quality rice raised the average import price above the import parity price for below-poverty-line rice. 19. For example, from 1995­96 through 2002­03, the value of manufactured jute products was more than three times the value of raw jute exports (calculated based on data in World Bank 2005). 20. The jute export tax equivalent in East Pakistan during the 10 years leading up to the independ- ence of Bangladesh from Pakistan in 1971 was at least as severe as and probably more severe than the jute export tax equivalent during the rest of the 1970s. 21. Pursell's estimates (2005) for 2002 are domestic production at 177,000 tons, official imports from India at 349,000 tons, official imports from other countries at 93,000 tons, and sugar smuggled from India at 400,000 tons. 22. Until June 2002, the Trading Corporation of Bangladesh and the Bangladesh Sugar and Food Industries Corporation, a public enterprise, shared a monopoly on sugar imports (Pursell 2005). 23. Another parastatal, the Bangladesh Water Development Board, was established in the early 1960s to implement large-scale projects in surface water irrigation, flood control, and drainage. The initiatives were based on recommendations contained in a master plan prepared in the aftermath of disastrous floods in the mid-1950s. 24. Despite privatization in the distribution of fertilizers, the public sector continued to account for most urea production. Bangladesh 335 25. Like fertilizer, irrigation water was also heavily subsidized in the early years of the expansion in irrigation. There was a rapid reduction in subsidies in the late 1970s when irrigation equipment previously owned by the public sector was privatized. Since 1988, the government has eliminated all restrictions on imports of irrigation equipment by the private sector, cancelled import duties on agri- cultural machinery, and removed restrictions aimed at standardization and quality control in the pro- duction of agricultural machinery. These policy changes have had a significant impact on the use of low-end irrigation equipment, especially shallow tube wells, and they have led to a rapid expansion in the total area under irrigation in the country (Ahmed 2001). 26. In most years, the country has produced urea from domestic natural gas. Net exports of urea on a large scale began in 1988 (Renfro 1992). 27. For an analysis of the oscillation of postcolonial governments between technocratic and pop- ulist policies, see Huntington and Nelson (1976). 28. In reforming the grain procurement and ration shop system in the early 1990s, the government was able to avoid direct conflict with consumer groups by only gradually reducing the price subsidy on ration shop grain. Without the support of consumer groups, which no longer reaped significant bene- fit from the ration shop system, millers lacked the political clout to resist the call to end the millgate procurement system (Chowdhury and Haggblade 2000). 29. Various studies of the comparative advantage of Bangladesh in agriculture demonstrate that the attainment of self-sufficiency in rice production is an important sociopolitical objective and also an eminently sensible goal strictly from the point of view of economics (see Mahmud, Rahman, and Zohir 2000; Shahabuddin 2000; Shahabuddin and Dorosh 2004). 30. Government interventions through domestic procurements were largely ineffective in main- taining floor prices for producers. A number of factors contributed to the failure of domestic procure- ment, including an inadequate coverage of production areas by procurement centers, cumbersome payment procedures that raised the transaction costs for small farmers, the lack of financial resources in the public food distribution system, and collusion between traders and officials that enabled traders to capture the margin between the market price and the procurement price (Shahabuddin 1996). 31. When the issue price of fertilizer was reduced in the 1994 budget, farmers expected that they would be the major beneficiaries. So, their notion of a just price was outraged when they were required to pay prices that were nearly double (and sometimes more) the prices the middlemen paid to the factories. References Abdullah, A. 1996. "Urea Market in Bangladesh, 1994­95: The Anatomy of a Crisis." In State, Market and Development: Essays in Honour of Rehman Sobhan, ed. A. Abdullah and A. R. Khan, chap. 12. Dhaka, Bangladesh: University Press. Abdullah, A., and Q. Shahabuddin. 1993. "Critical Issues in Bangladesh Agriculture: Policy Response and Unfinished Agenda." Paper presented at the Asian Development Bank and Academy for Planning and Development conference, "Bangladesh Economy in Transition," Dhaka, Bangladesh. Ahmed, A. W. Nuruddin, L. H. Chowdhury, and S. Haggblade. 2000."History of Public Food Interven- tions in Bangladesh." In Out of the Shadow of Famine: Evolving Food Markets and Food Policy in Bangladesh, ed. R. Ahmed, S. Haggblade, and T.-E. Chowdhury, 121­36. Baltimore: Johns Hopkins University Press; Washington, DC: International Food Policy Research Institute. Ahmed, N., Z. Bakht, P. A. Dorosh, and Q. Shahabuddin. 2007. "Distortions to Agricultural Incentives in Bangladesh." Agricultural Distortions Working Paper 32, World Bank, Washington, DC. Ahmed, R. 2001. Retrospect and Prospects of the Rice Economy of Bangladesh. Dhaka, Bangladesh: University Press. Ahmed, R., S. Haggblade, and T.-E. Chowdhury, eds. 2000. Out of the Shadow of Famine: Evolving Food Markets and Food Policy in Bangladesh. Baltimore: Johns Hopkins University Press; Washington, DC: International Food Policy Research Institute. 336 Distortions to Agricultural Incentives in Asia Anderson, K., M. Kurzweil, W. Martin, D. Sandri, and E. Valenzuela. 2008. "Measuring Distortions to Agricultural Incentives, Revisited." World Trade Review 7 (4): 675­704. Bakht, Z. 1999. "Impact of SAFTA on the Official and Unofficial Trade of Bangladesh." Unpublished working paper, Bangladesh Institute of Development Studies, Dhaka, Bangladesh. ------. 2001a."Preparation of the Sixth Five-Year Plan: Background Paper on Industry." Unpublished working paper, Bangladesh Institute of Development Studies, Dhaka, Bangladesh. ------. 2001b. "Trade Liberalization, Exports and Growth of Manufacturing Industries in Bangladesh." In Strategies for Industrialization: The Case of Bangladesh, ed. M. M. Huq and J. Love, chap. 5. Dhaka, Bangladesh: University Press. BBS (Bangladesh Bureau of Statistics). various years. Statistical Yearbook of Bangladesh. Dhaka, Bangladesh: BBS. Chowdhury, T.-E., and S. Haggblade. 2000. "Dynamics and Politics of Policy Change." In Out of the Shadow of Famine: Evolving Food Markets and Food Policy in Bangladesh, ed. R. Ahmed, S. Haggblade, and T.-E. Chowdhury, 165­88. Baltimore: Johns Hopkins University Press; Washington, DC: International Food Policy Research Institute. del Ninno, C., and P. A. Dorosh. 2003. "Impacts of In-Kind Transfers on Household Food Consump- tion: Evidence from Targeted Food Programs in Bangladesh." Journal of Development Studies 40 (1): 48­78. del Ninno, C., P. A. Dorosh, and N. Islam. 2002. "Reducing Vulnerability to Natural Disasters: Lessons from the 1998 Floods in Bangladesh." IDS Bulletin 33 (4): 98­107. Dorosh, P. A. 2001. "Trade Liberalization and National Food Security: Rice Trade between Bangladesh and India." World Development 29 (4): 673­89. ------. 2006. "Accelerating Income Growth in Rural Bangladesh." Background report, World Bank, Washington, DC. Dorosh, P. A., N. Farid, R. Amin, and M. A. Aziz. 2004. "Policy Response to Production Shocks: The 1997/98 Aman Shortfall and the 1998 Flood." In The 1998 Floods and Beyond: Towards Comprehen- sive Food Security in Bangladesh, ed. P. A. Dorosh, C. del Ninno, and Q. Shahabuddin, 155­80. Dhaka, Bangladesh: University Press; Washington, DC: International Food Policy Research Institute. Dorosh, P. A., and S. Haggblade. 1997. "Shifting Sands: The Changing Case for Monetizing Project Food Aid in Bangladesh." World Development 25 (12): 2093­104. Dorosh, P. A., and Q. Shahabuddin. 2005. "Trade Liberalization and Food Security in Bangladesh." In Economic Reforms and Food Security: The Impact of Trade and Technology in South Asia, ed. S. C. Babu and A. Gulati, 141­62. New York: Haworth Press; Washington, DC: International Food Policy Research Institute. FAOSTAT Database. Food and Agriculture Organization of the United Nations. http://faostat.fao.org/ default.aspx (accessed May 2008). Global Development Network Growth Database. Development Research Institute, New York University. http://www.nyu.edu/fas/institute/dri/global%20development%20network%20growth%20database .htm (accessed June 2007). Grindle, M. S. 1991."The New Political Economy: Positive Economics and Negative Politics." In Politics and Policy Making in Developing Countries: Perspectives on the New Political Economy, ed. G. M. Meier, 41­68. Richmond, CA: ICS Press. Hossain, M. 2001. "Recent Development and Structural Changes in Bangladesh Agriculture: Issues for Reviewing Strategies and Policies." In Changes and Challenges: A Review of Bangladesh's Develop- ment 2000, ed. R. Sobhan, chap. 3. Dhaka, Bangladesh: University Press and Center for Policy Dialogue. Huntington, S., and J. Nelson. 1976. No Easy Choice, Cambridge MA: Harvard University Press. Mahmud, W., S. H. Rahman, and S. Zohir. 2000. "Agricultural Diversification: A Strategic Factor for Growth." In Out of the Shadow of Famine: Evolving Food Markets and Food Policy in Bangladesh, ed. R. Ahmed, S. Haggblade, and T.-E. Chowdhury, 232­60. Baltimore: Johns Hopkins University Press; Washington, DC: International Food Policy Research Institute. Bangladesh 337 Ministry of Finance. 2005. Bangladesh Economic Review. Dhaka, Bangladesh: Ministry of Finance. Osmani, S. R., and A. Quasem. 1990. "Pricing and Subsidy Policies for Bangladesh Agriculture." Unpublished working paper, Bangladesh Institute of Development Studies, Dhaka, Bangladesh. Pursell, G. 2005. "Free Trade between India and Bangladesh: A Case Study of the Sugar Industry." Unpublished report, World Bank, Washington, DC. Rahman, S. H. 1992. "Trade Policies and Industrialization in Bangladesh: An Assessment." Unpub- lished working paper, Bangladesh Institute of Development Studies, Dhaka, Bangladesh. ------. 1994. "The Impact of Trade and Exchange Rate Policies on Economic Incentives in Bangladesh Agriculture." Working Papers on Food Policy in Bangladesh 8, International Food Policy Research Institute, Washington, DC. Ravallion, M. 1990. Markets and Famines. Dhaka, Bangladesh: University Press. Renfro, R. Z. 1992. "Fertilizer Price and Subsidy Policies in Bangladesh." World Development 20 (3):437­55. Shahabuddin, Q. 1996. "Public Interventions in Food Grain Markets in Bangladesh: An Appraisal." In State, Market and Development: Essays in Honour of Rehman Sobhan, ed. A. Abdullah and A. R. Khan, chap. 13. Dhaka, Bangladesh: University Press. ------. 2000. "Comparative Advantage in Bangladesh Agriculture." Bangladesh Development Studies 26 (1): 37­68. Shahabuddin, Q., and P. A. Dorosh. 2004. "Diversification of Agricultural Production: Comparative Advantage in Crops and Implications of the World Trade Organization." In The 1998 Floods and Beyond: Towards Comprehensive Food Security in Bangladesh, ed. P. A. Dorosh, C. del Ninno, and Q. Shahabuddin, 313­34. Dhaka, Bangladesh: University Press; Washington, DC: International Food Policy Research Institute. USAID (United States Agency for International Development) 1982. "Pricing, Subsidies and Related Policies in Food and Agriculture." Policy Paper, USAID, Washington, DC. USDA (United States Department of Agriculture). 2001. "India: Grain and Feed Annual Report 2001." Report, Foreign Agricultural Service, USDA, Washington, DC. Wolf, E. 1971. "Peasant Wars of the Twentieth Century." London: Faber and Faber. World Bank. 1979. Bangladesh Food Policy Issues. Report 2761-BD. Washington, DC: World Bank. ------. 1994. Bangladesh: From Stabilization to Growth. Report 12724-BD. Washington, DC: World Bank. ------. 2004. An Overview. Vol. 2 of Trade Policies in South Asia: An Overview. Report 29949. Washington, DC: World Bank. ------. 2005. Bangladesh: Growth and Export Competitiveness. Report 31394-BD. Washington, DC: World Bank. 10 INDIA Garry Pursell, Ashok Gulati, and Kanupriya Gupta This chapter analyzes the impacts on price incentives in India's agricultural sector that have resulted from government price, trade, and exchange rate policies. Previ- ous studies have established that, from the 1970s to 1995, the incentive system strongly favored manufacturing and services over the principal agricultural crops, although the extent of the antiagricultural bias had diminished considerably by the mid-1990s (Pursell 1999).1 Our study updates these earlier estimates to 2005 and incorporates estimates for agriculture going back to 1965. In addition, we extend the previous work by considering policies affecting the incentives for the produc- tion of fresh fruits and vegetables and for dairying, which account for large shares of the rural economy as measured by the respective contributions to gross domes- tic product (GDP). We also briefly discuss policies affecting food processing. The chapter is organized as follows. In the next section, we provide overviews of economic growth and structural change in the Indian economy and the evolu- tion of trade policies since independence, including the effects of current govern- ment policies on agricultural trade with the country's neighbors in South Asia. In the subsequent section, we discuss the exchange rate and its interaction with trade policies, especially during the massive rupee devaluation of 1985­93. We then offer quantitative evidence on nominal rates of assistance (NRAs) in various agri- cultural subsectors, including subsidies for electricity and fertilizer inputs. We also estimate the distortions in the incentives for farmers relative to the incentives for The authors thank the members of the central project team (especially Marianne Kurzweil and Ernesto Valenzuela) for their contributions and technical help with the data spreadsheets from which several of our tables and figures have been generated. 339 340 Distortions to Agricultural Incentives in Asia producers of nonagricultural tradables. We use this evidence to show that the ini- tial, notable antiagricultural bias of government policies gradually gave way to a slight bias in favor of agriculture. Finally, we discuss the political economy forces that are likely to influence the direction of government policies, including the possibility that a strong proagricultural bias may emerge as has occurred in more- advanced, densely populated economies in East Asia and elsewhere. Economic Growth and Structural Change Agriculture accounted for slightly more than 50 percent of GDP in 1950, but its share is now less than 20 percent.2 Most of this decline occurred in crop agricul- ture. In contrast, the livestock subsector (mainly dairying) grew more rapidly than the rest of the economy. The GDP share of this subsector rose from 2 percent in the early 1980s to 4.4 percent in 2005. Forestry was important after independence, but, in 2003, it provided only around 1 percent of GDP, as did fishing, despite fairly speedy growth over a long period, mainly impelled by fish farm develop- ment. Meanwhile, mining's contribution grew from less than 1 percent in the 1950s to 2.8 percent in 2005. The steep decline in agriculture's contribution to the economy since independence has been generated principally by the more rapid growth in the services sector, from barely 30 percent of GDP in 1950 to 63 percent in 2003. The growth of the services sector has been especially brisk since the early 1990s, and it accelerated during the late 1990s and early years of the next decade because of swift advances in information technology and export outsourcing. The share of manufacturing in the economy also increased, but not nearly as quickly as it did in many other developing countries. The share of manufacturing in GDP in 2005 was 16 percent (much lower than in East Asian economies). This compares with 10 percent in 1950, shortly after independence (table 10.1). The contribution of the rural sector to GDP has declined by almost 70 percent since independence. However, the sector still accounted for around 60 percent of national employment in 2003, and this share had been declining only slowly since the early 1990s (partly because of slow growth in manufacturing employment). International trade in agricultural products has always been tiny relative to the size of the agricultural sector. In 2005, agricultural imports were equivalent to only 0.8 percent of the value of agricultural production (or 2.1 percent if edible oil imports are included), while exports were 6.4 percent of production. Even during earlier periods, when agriculture represented a major share of total trade (for example, 27 and 44 percent of total imports and exports, respectively, in 1960), imports and exports still only accounted for a little more than 3 percent of agri- cultural production (table 10.2). The most important primary agricultural import initially was wheat, followed by cotton and powdered milk. Beginning around the mid-1970s, the need for Table 10.1. Sectoral Share of GDP and Employment, India, 1950­2005 (percent) Sector 1950 1960 1970 1980 1990 2000 2003 2004 2005 All agriculturea 56.9 46.7 46.1 38.9 31.3 23.4 20.9 18.8 18.3 Livestock -- -- -- 1.9 4.7 5.4 4.6 4.8 4.4 Crops -- -- -- 33.7 23.8 15.8 14.4 12.2 12.1 Forestry and logging 3.0 2.2 2.1 2.5 1.8 1.0 0.9 0.8 0.7 Fishing 0.4 0.5 0.6 0.7 1.0 1.1 1.1 1.0 1.1 Mining and quarrying 0.8 1.1 1.1 1.7 2.7 2.4 2.5 3.0 2.8 Manufacturing 10.6 13.8 13.8 16.3 17.1 15.6 15.2 15.9 16.0 Services 31.8 38.4 39.1 43.1 48.8 58.7 61.3 62.4 62.9 Agriculture's share in employment -- -- -- -- 69 62 60 -- -- Sources: Author computations based on Ministry of Finance, various; CGA Database 2008. Note: -- no data are available. a. Agriculture includes crop agriculture, horticulture, livestock activities, inland and ocean fisheries, and forestry activities. 341 342 Table 10.2. Profile of Agricultural and Manufacturing Imports and Exports, India, 1960­2005 (percent) Indicator 1960 1970 1980 1990 2000 2004 2005 Share in total imports Primary agricultural, fishing, and forest products 27.2 23.7 3.5 1.9 0.7 0.9 0.8 Edible oils 0.3 2.4 5.6 0.8 2.6 2.2 1.4 Manufactured products 66.1 61.3 42.9 56.9 51.9 46.2 47.1 Share in total exports Primary agricultural, fishing, and forest products 44.3 30.7 30.2 17.8 12.8 9.8 9.6 Manufactured products 45.3 53.6 59.2 72.3 77.7 73.0 70.3 Imports as a share of production Primary agricultural, fishing, and forest products 3.4 1.7 0.7 0.4 0.3 0.8 0.8 Edible oils 0.0 0.2 1.2 0.2 1.2 1.8 1.3 Manufactured productsa 9.5 4.2 5.6 6.3 8.6 11.6 13.3 Exports as a share of production Primary agricultural, fishing, and forest products 3.2 2.1 3.3 3.2 5.1 5.9 6.4 Manufactured products 3.7 3.4 4.1 6.1 11.4 13.8 14.3 Sources: Author computations based on Ministry of Finance, various; CGA Database 2008. a. Excludes petroleum, oils, lubricants, and nonmetallic mineral manufactures, which consist mainly of rough diamonds. India 343 these and other agricultural imports were eliminated because of growing domes- tic production, and, since then, they have accounted for only a small share of total imports even when they were imported during poor crop seasons, for example, in wheat and sugar. The largest consistent imports of processed food have involved edible oils. Imports of these products expanded rapidly during the 1970s and early 1980s, triggering a major government program to replace the imports through domestic production. For a while, edible oil imports declined, but, despite high tariffs (an 80 percent tariff on palm oil in 2006, for instance), import growth resumed in the 1990s, and, by 2005, imports accounted for about 40 percent of the domestic con- sumption of these products. During the 1950s and 1960s, agricultural products represented slightly less than half of the country's total merchandise exports, but the share has steadily declined since then and is now around 10 percent. A diverse range of agricultural products are regularly exported, including fish and fish preparations, oil cakes, cashew kernels, tea, coffee, tobacco, spices, fruits, vegetables, pulses, basmati rice, and, periodically, large quantities of sugar and common rice (more than 4 million tons of rice in 2004/05, for example). Except for cashew kernels, the share of exports in the total domestic production of these individual products is small. Since the late 1980s, manufactured exports have usually accounted for 70­80 percent of total merchandise exports, compared with 40­50 percent in the 1950s and 1960s. Manufactured exports have also increased in relation to total manufac- turing production, especially since the late 1980s, growing about twice as quickly, at 20­30 percent a year, in 2000­05. However, relative to many other developing countries, these exports still constitute a small share of manufacturing sector out- put (around 15 percent in 2005). Service sector exports have also been growing rapidly in recent years, at about 25 percent annually. In 2004, net invisible exports represented 4.9 percent of GDP, compared with the negative net balances before 1990. The most dynamic compo- nents of this trade are software exports, other exports related to information tech- nology, and the outsourcing of services. The share of food expenditure in household budgets is high in India, amount- ing to 54 percent and 42 percent of total per capita consumption expenditure in rural and urban areas, respectively. The shares of food in the budgets of the poor- est 10 percent of households is even higher, averaging 62 percent in rural areas and 58 percent in urban areas in 2003. Because of these high shares of food in household budgets (the shares have been declining only slowly), it is not surpris- ing that food prices and food availability are regular news items in the local and national press and other media, and that food prices, especially sudden increases such as in 2008, are highly sensitive politically. One of the most prominent 344 Distortions to Agricultural Incentives in Asia objectives of the independence movement in India was to establish institutions and policies that would permanently eliminate catastrophic famines, such as those that occurred during the colonial period, and that would also ensure that basic food items would be available to the entire population at affordable prices. To achieve these food policy goals, the government has intervened in food grain markets since the late 1940s. In 1958, it established the current public distribution system, which sells basic foods at subsidized prices through fair price shops. (There are currently about 460,000 of these shops.) For most of its history, the system has distributed wheat, rice, sugar, and edible oils on the basis of ration cards that enti- tle the bearers to specified quantities of food items at announced low prices. In June 1997, the system was changed by distinguishing, according to fixed criteria, between below-poverty-line and above-poverty-line buyers. The former group was eligible to benefit from especially low prices, while the latter group was eligible to buy the food items at prices that were slightly below free-market prices. In 2001, edible oils were removed from the system, and, in 2002, the role of sugar was dras- tically reduced by allowing subsidized sales only to below-poverty-line households. The principal government food subsidy activity today is the sale of rice and wheat through the fair price shops to below-poverty-line households and distribution as part of other antipoverty programs. In 2003, the total government food subsidy was estimated at Rs 258 billion (about US$5.7 billion, or 0.83 percent of GDP); this subsidy was defined as the Food Corporation of India's total procurement han- dling and distribution costs, less the subsidized sales value. Trade and Exchange Rate Policies since Independence During the second half of the 1800s and until about 1921, the British rulers of India followed free trade policies that imposed few restrictions or taxes on exports or imports.3 These free trade policies began to change in the early 1920s following the collapse of the post­World War I boom, and protective tariffs continued to be introduced during the 1920s and 1930s. Then, in 1940, general controls were imposed on all imports and exports to deal with scarcities in goods, shipping, and foreign exchange and to address wartime priorities. The general rule was that imports would only be allowed if they were essential and could not be supplied by local industries. After independence in August 1947, the government relaxed wartime import controls by expanding the scope of lists of goods that could be imported without obtaining a license (the open general license) and by raising tariffs to take pressure off the import licensing system. However, the start of the Second Five-Year Plan in 1956 coincided with a severe foreign exchange crisis, and, until 1966, the compre- hensive administration of the import licensing system was tight. These foreign India 345 trade policies were an extension of more general economic policies under which the commanding heights of the industrial economy (which excluded farming) were dominated by state enterprises, and the private sector (including agriculture) was subject to extensive controls. Collectively, these controls came to be known as the license Raj (the Hindi word for reign or rule). In June 1966, the rupee was devalued. This was accompanied by a brief episode of liberalization during which import licensing was relaxed, tariffs were cut, and export subsidies were abolished or reduced. However, the import licensing system remained intact, and, by 1968, most liberalizing initiatives had been reversed and tight import and domestic controls had been reinstated. This remained the situa- tion until the end of the 1970s, when a new phase of slow, partial liberalization commenced. The slow liberalization trend of the late 1970s and early 1980s included the relaxation of industrial licensing rules, regular additions of intermediate products and capital goods to open general license lists, and tariff increases that succeeded in capturing some of the economic rents inherent in the import licensing system. Liberalization gained momentum during the Rajiv Gandhi government adminis- tration (1985­89) when the economic growth rate accelerated above the slow average rate that had characterized the previous 40 years. The boom culminated in a severe foreign exchange crisis in 1991. The government reacted by imple- menting a sharp devaluation. The devaluation was accompanied by sweeping liberalizing measures that removed many of the key domestic controls over man- ufacturing, nearly all the import licensing system for intermediate and capital goods, and a key export subsidy program. The initiatives also included a four-year program of tariff reductions. These reductions continued into the 1990s and brought tariffs down far below the extremely high and prohibitive rates of the 1980s (when they averaged more than 100 percent), though the tariffs were still high by international standards. Domestic policies and trade policies that affected the rural sector were basi- cally untouched by the 1991 reforms. In particular, government enterprises con- tinued to dominate the domestic and international trade in cereals (notably the Food Corporation of India, which periodically imported wheat to meet domestic shortages), and agricultural products remained subject to the import licensing system that applied to all consumer goods. With some important exceptions, import licenses were not issued for agricultural products. The system thus amounted to an import ban on agriculture. The exceptions in agriculture included a few products for which imports had been open even during the most restrictive periods (pulses, for example) and other products, such as cotton and wool, for which unrestricted low-tariff imports of important inputs had been successfully negotiated by influential industrial lobbyists. Edible oils were also 346 Distortions to Agricultural Incentives in Asia imported on a large scale (despite efforts undertaken to replace the imports through domestic production), first, by state trading firms holding import monopoly rights and, later, by the private sector. However, the edible oil imports were subject to high tariffs. It has been estimated that, in the mid-1990s, five years after the 1991 reforms, about two-thirds of tradable GDP was still protected by explicit nontariff barriers (about 36 percent in manufacturing, 84 percent in agriculture, and 40 percent in mining). During the second half of the 1990s, this situation began to change, in large measure in response to international pressures linked to the Uruguay Round agreements and the negotiations associated with them. Starting in 1998, the gen- eral import licensing system began gradually to be dismantled, and, on April 1, 2001, the last 715 of 2,714 tariff lines (including nearly all the agricultural tariff lines) were removed, and the system itself was abolished. Understandably, after almost 50 years of de facto autarchy, the lifting of these controls generated considerable apprehension over the ability of domestic pro- ducers of manufactured consumer goods and agricultural products to compete with imports. A war room was created at the Ministry of Commerce, and a list of 300 sensitive products was drawn up so that imports of these products could be monitored to ensure that prompt action would be taken to preempt or minimize disruptions in local production. More substantively, during and following the Uruguay Round negotiations, the government sought to protect or subsidize domestic producers by marshaling all the techniques not explicitly forbidden by the World Trade Organization agreements. They included techniques of uncertain legality that the government believed might be used without attracting serious complaints from trading partners.4 It became apparent soon after the final abandonment of import licensing in April 2001 that the war room psychology had greatly exaggerated the danger repre- sented by rapidly expanding imports. During the next couple of years, existing tar- iffs and the other measures that had been introduced proved more than adequate in keeping out competing manufacturing and agricultural imports. Eventually, without a formal announcement, the sensitive products lists quietly disappeared from official publications and public discussion. At the same time, manufactured exports entered a new phase of rapid expansion that continued into 2007 (at around 20­25 percent annually). This was supplemented by similarly rapid growth in exports in services. Together with greater capital inflows, these developments helped create a strong balance of payments and historically high foreign exchange reserves, and they were accompanied by brisk economic growth. Responding to the new confidence that these changes created, the government commenced a program of drastic reductions in industrial tariffs in April 2003 that, over the next four years, lowered the average tariff by approximately India 347 Figure 10.1. Import Tariffs, Agricultural and Nonagricultural Products, India, 2002­06 50 45 40 % 35 average 30 tariff, 25 20 15 import unweighted 10 5 0 2002 2003 2004 2005 2006 year all tariff lines agriculture, Harmonized Commodity Description and Coding System 1­24 nonagriculture, Harmonized Commodity Description and Coding System 25­99 Source: Author computation based on data in Goyal, various. Note: 2002/03 (2003) and 2003/04 (2004) include para-tariffs that were abolished in 2004/05 (2005). The averages reflect ad valorem tariffs only. They do not take account of specific components in compound duties. two-thirds, from over 33 percent to about 12 percent (figure 10.1).5 This was followed by another tariff reduction in the 2007 budget. Because of these cuts, India became one of the world's low-protection countries (measured according to average ad valorem industrial tariffs) after having been one of the world's most protected countries.6 The average industrial tariffs are now slightly higher than the corresponding average in China and the Republic of Korea and about the same as the average in Sri Lanka, which has traditionally been considered the sole low-protection economy in South Asia. Moreover, because of the top-down reduction process, the industrial tariff structure is uniform. Thus, in 2007, over 80 percent of the industrial tariff lines were at or below the new general maximum of 10 percent; this limited the scope for high effective protection through escalated tariff structures. However, from the beginning, agriculture and processed foods were not cov- ered by the new tariff reduction program (figure 10.1). In 2006, the unweighted average tariffs protecting these industries (Harmonized Commodity Description 348 Distortions to Agricultural Incentives in Asia and Coding System, 1­24) were about 40 percent, almost four times the average industrial tariffs and among the highest in the world. Moreover, these tariffs are highly dispersed; about 15 percent of the tariffs are in the 50­100 percent range. These high tariffs have been maintained despite substantial tariff redundancy among most agricultural commodities (see elsewhere below). The domestic prices for many commodities are not only lower than duty-inclusive import prices, but are also frequently lower than duty-exclusive import prices. This special treatment provided through agricultural trade policies reflects the pressure exerted by many farmer and processor interest groups and mediated through and supported by the Ministry of Agriculture. Regional trade agreements As a by-product of the country's highly protective agricultural trade policies, the trade in primary and processed agricultural and livestock products between India and its South Asian neighbors has been seriously hindered.7 Ironically, the main victim has likely been Indian exports to these countries rather than Indian imports from the region. Indian domestic prices are low, and, even under a free trade regime, the products of neighboring countries would face difficulties com- peting there. Meanwhile, India's high tariffs and other barriers to imports have reinforced and helped justify domestically the reluctance of other South Asian countries to reduce their own barriers to agricultural trade either multilaterally or as part of regional and bilateral preferential trade arrangements. Thus, for many years, the South Asian Preferential Trade Agreement had only limited relevance in regional agricultural trade because all agricultural imports to India were subject to that country's discretionary import licensing system. Import licensing was lifted among members of the South Asian Preferential Trade Agree- ment in 1998, three years before it was finally phased out by this group with respect to the rest of the world. Nonetheless, since then, the combination of high redundant tariffs, low domestic prices for most agricultural products, and the continuing role of state trading enterprises has meant that it is impossible for these countries to compete in the majority of India's domestic markets. Except for Bhutan and Nepal, the government of India has provided few tariff prefer- ences for agricultural products under its bilateral trade agreements, notably, in its bilateral free trade agreement with the government of Sri Lanka. Under the South Asia Free Trade Agreement, most agricultural tariff lines, including the lines for processed foods, are on the government of India's sensitive list of the products for which it has made no commitments. The reluctance of the government to provide tariff preferences partly reflects a desire to prevent imports from third countries through preference-receiving India 349 countries. Reciprocally, Bangladesh, Pakistan, and Sri Lanka have been unwilling to provide tariff preferences for agricultural products under the South Asian Pref- erential Trade Agreement and have also placed most agricultural products that they produce on their own South Asia Free Trade Agreement sensitive products lists. This is largely a reaction to the Indian position because, even though these agreements, in principle, involve a number of countries, the potential regional trade that really matters is, in practice, the bilateral trade with India. However, in Bangladesh and Sri Lanka, the reciprocity also reflects a realistic assessment that agricultural free trade with India would generate more agricultural imports from India than exports to India, in the process threatening the viability of some of the more highly protected agricultural industries of these countries, such as sugar, various fresh fruits, vegetables, and a wide range of processed foods in Bangladesh, and, in Sri Lanka, rice, potatoes, and onions.8 If it were opened up, India-Pakistan bilateral agricultural trade would probably be more balanced. Recent studies suggest that there is considerable potential for welfare-improving bilateral trade in wheat and sugar. The direction and timing of this trade would vary with weather and other seasonal factors and, compared to trade through the ports with countries outside the South Asia region, would benefit from large transport cost savings, especially trade between the Indian northwest states and Pakistan Punjab. It is also likely that the poultry and other livestock products of Pakistan could be profitably exported to India. However, bilateral trade between India and Pakistan is hostage to the difficult political relationship of the two countries, which is reflected in Pakistan's positive list of products that may be legally imported from India. This list includes almost no agricultural products. Moreover, rules enforced in both countries (with a few minor exceptions) do not allow trade over the land border. The exchange rate regime and trade policies Trade policies must be understood in the context of exchange rate policies.9 The rupee was pegged to the pound sterling before independence in 1947, and this continued until September 1975. In June 1966, the fixed rupee-pound rate was devalued by 57.5 percent and by the same proportion to the U.S. dollar, to Rs 7.50. In line with the fixed link with the pound, the rupee-dollar rate floated down slightly between 1966 and 1975. In September 1975, the peg to the pound sterling was removed, and, until 1992, the rate was fixed by the Reserve Bank of India. In 1992, this system was replaced by a managed floating rate, whereby the Reserve Bank of India allowed the rupee to move in relation to a basket of currencies. A prolonged period of nominal devaluation began in 1982. The devaluation rate was rapid at first. It slowed for three years in the mid-1980s and then accelerated 350 Distortions to Agricultural Incentives in Asia between 1989 and 1991. This process culminated in a major devaluation to deal with the July 1991 foreign exchange crisis, and the nominal rupee-dollar rate fell by approximately 70 percent between 1991 and 1993. During the next 10 years, the exchange rate was devalued at a modest pace that approximately offset domestic inflation in relation to inflation among India's principal trading partners. From 2003 to mid-2008, the nominal rupee-dollar rate appreciated because of a new export boom and the buildup of large foreign exchange reserves. The economic significance of the trends in the country's nominal exchange rates becomes clearer if one takes account of the nominal rates in relation to the currencies of India's principal trading partners (not only the U.S. dollar) and of the inflation rate in India relative to the inflation rates in these countries. These effects are systematically captured in real effective exchange rate series. Compre- hensive real effective exchange rate indexes are available for India since 1980. Fig- ure 10.2 illustrates an index for this period weighted by India's total trade with 25 countries, and this index has been linked to earlier estimates of the World Bank and others to produce a series starting in 1965. Based on this series, the interaction between India's exchange rate history and the country's trade policies may be Figure 10.2. Index, Real Effective Exchange Rate, India, 1965­2004 275 250 100 = 225 1980 200 175 rate, 150 125 exchange 100 75 50 effective 25 real 0 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 year real effective exchange rate (total trade weights, 25 countries), Indian fiscal year